Intelligent and Transformative Production in Pandemic Times

To assure appearance quality of industrial products, appearance inspection by human vision is implemented in many manufacturing industries. In appearance inspection by human vision, small visual defects such as dirt, scratches, surface dents, and unevenness of the coating color are inspected. In actual defects in appearance inspection processes, the visibility differs depending on the positional relationship between the inspector and the defect. Furthermore, visibility varies depending on the type and characteristic of defects. Therefore, to comprehensively inspect these defects, it is required to change the positional relationship between the inspector and defects by appropriate body movements of the inspector. In particular, in case of product size is relatively large such as a body, glass of an automobile, there is strongly required the body movement of the inspector to realize positional relationship in which defects can be detected. However, no studies have been conducted on appropriate body movements based on the defect positional relationship between the inspector and defects, in actual appearance inspection processes, work training is only carried out by on the job training based on experience of skilled inspectors. Therefore, this study focused on appearance inspection that required appropriate body movements of inspectors during the inspection, and also developed a work training system for appearance inspection using motion capture and large size display. Moreover, to show how to utilize the developed training system, the experiment using the general work procedure manual and the work procedure manual that teaches detailed body movements according to size of products and height of an inspector are designed as experimental factors, and their effects on defect detection are evaluated with 12 subjects. As a result, it is quantitatively obtained that the latter work procedure manual has the effect of improving the defect detection rate about 5.9%. In particular, it is also obtained to be effective in case of detecting small luminance contrast and detectable range of defects. From the above, it is considered that the work training methodology including body movements is effective, and it is shown that the developed training system may contribute to work design and work training in the actual appearance inspection processes.

Crowdsouring with cooperative collaboration among experts should be performed by teams that consider compatibility among crowd workers. In this study, we develop a system that can automatically form an organization to efficiently perform complex and large-scale projects in crowdsourcing. The system targets projects that can be divided into multiple tasks, and the output organization consists of subgroups and subleaders that correspond to the tasks. The compatibility among crowd workers is calculated by weighted social network generated with the past experience of collaboration. Furthermore, we develop an algorithm based on the greedy method to search teams with compatibility optimization under the constraints of budget and skills using the social network. In the experiment, we collected 169,627 users on 33,983 repositories from GitHub, and determined the social network and skills of the workers by the programming languages, topics, and user contributions from the repositories. The characteristics of the developed system are observed by the simulation experiments with a virtual project with skills and budget requirements. In particular, the comparison of teams formed with minimum budget suggests that the proposed team formation algorithm achieves a 96% higher compatibility among crowd workers than the algorithms that do not consider compatibility. Furthermore, the proposed algorithm was compared with the algorithm of the full search in related work, and it was found that full search algorithm enhances the team compatibility when a single team is formed under strict budget constraints.

Recently, companies have been shifting from lean manufacturing to cell manufacturing using industrial robots. Nof et al. have proposed cooperation requirement planning (CRP) among multiple industrial robots and tasks in the production process. CRP comprises two parts, CRP-I and CRP-II. CRP-I establishes the conditions for ensuring cooperation among multiple industrial robots, whereas CRP-II formulates an operation plan for optimizing the manufacturing objectives using the conditions established by CRP-I. Nof’s study describes a method for formulating a consistent operation plan based on a two-step method procedure where conditions are set by CRP-I and optimized by CPR-II in multiple industrial robots. Furthermore, the matrix approach is a method of representing the relationship between inputs and outputs and examining operational plans. The present study focuses on CRP with multiple industrial robots and proposes a mathematical model using a matrix approach (Matsui’s equation) that includes scheduling method. The proposed mathematical model demonstrates the possibility of reducing the abovementioned two operation steps (CRP-I and CRP-II) to one and includes the scheduling theory by adding one additional matrix to Matsui’s equation. Furthermore, the mathematical model can compare the corrected values obtained for two cases involving CRP with multiple industrial robots based on this modified matrix approach with the standard values obtained from the original approach. The matrix approach examines the operation plans of multiple industrial robots by comparing the standard values and corrected values. This paper presents a mathematical model and development from the conventional two-stage method (CRP-I and CRP-II) to a one-stage method (matrix approach) of Matsui based on the cooperation requirement planning of multiple industrial robots.

In recent years, global warming has become a serious problem in a global supply chain which is a series of cross-border transaction including custom duty that is a tax imposed for imported goods. In order to prevent the global warming, Green House Gas (GHG) emissions needs to be reduced throughout the supply chain. Moreover, procurement costs, material-based GHG emissions, tariff in countries are different by each country. In addition, the Trans-Pacific Partnership (TPP), which is a free trade agreement signed between 11 countries, including Japan and Malaysia, have promoted the trade in parts and products among the TPP participant countries without customs duty. Thus, the network on the global supply chain affects not only costs but also GHG emissions. On the other hand, the disruption by COVID-19 caused adverse impacts to redesign supply chains all over the world, where parts or materials are not provided from current suppliers by the disruption. Thus, the network may be reconfigured, which brings different GHG emissions in the global supply chain network before and after the disruption across TPP countries. The purpose of this study is to evaluate material-based GHG emissions under COVID-19 disruption on redesigning global supply chain network across TPP countries. Firstly, global supply chain network is modeled and formulated. Next, numerical experiments are conducted for evaluating material-based GHG emissions under disruption scenarios. Finally, the results are analyzed in terms of GHG emissions and costs. The result shows that the highest reduction ratio of the total GHG emission on a global supply chain is 58.4% compared to the baseline.

In recent years, natural disasters have occurred worldwide, causing tremendous damage. It is difficult to predict the scale of damage caused by natural disasters before they occur; and even after the scale of damage is understood, it is difficult to make accurate recovery plans and decisions based on limited information. Therefore, it is necessary to construct a system that supports decision-makers. Emergency logistics is a new field of logistics that plays an important role in disaster relief and recovery. The inventory routing problem (IRP) in emergency logistics has been attracting attention to reduce damage caused by disasters. Unlike usual logistics, emergency logistics considers two uncertainties: the uncertainty of the demand scale and the uncertainty of the distribution network, such as roads. Oshima et al. (2020) proposed an IRP model that focuses on inventory management in secondary storage warehouses where processing operations are possible, in addition to demand points, assuming the delivery plans for materials such as lumber are planned in a situation where available distribution networks are known or have been restored. On the other hand, no model has been proposed that considers damage to distribution networks, such as roads, in the recovery phase. In this study, we propose an IRP model that considers the uncertainty of the road network and utilizes existing facilities as processing bases. In addition, numerical experiments were used to verify the effects of using processing bases in emergency logistics for machinable building components.

Nowadays, the economic activities cause some environmental problems for not only the global warming by Greenhouse Gas (GHG), but also the material starvation by mass consumption of resources. In order to resolve their problems simultaneously, Japanese government took effect with the plastic regulation and announced that GHG emissions became zero by 2050. In order for manufacturing companies to develop sustainably, they are required to design the Global Closed-Loop Supply Chain (GCLSC) network to recycle End-of-Life (EOL) products and reduce GHG emissions. The GCLSC network integrates the global supply chain network, which is a series of cross-border transaction of product, and the local reverse supply chain network, which is a domestic transaction for collecting and recycling EOL products. However, it is necessary to select appropriate global suppliers considering GHG emissions because the amount of material-based GHG emissions in EOL product is dependent on manufacturing country. On the other hand, the GCLSC network is costly for procurement, manufacturing, transportation, recycling and opening facilities costs. Thus, the decision maker who designs designing GCLSC network should consider not only cost, but also environmental aspects such as recycling EOL product and GHG saving weight simultaneously. This study designs the GCLSC network to minimize total cost and to maximize entire recycling rate on global and local supply chain network. Furthermore, the GHG emission from components procurement and the GHG saving weight by recycling EOL product is evaluated. First, the GCLSC network is modeled. Next, the objective function for minimizing total cost and maximizing entire recycling rate are formulated with integer programming. Finally, a numerical experiment is conducted and evaluated in terms of costs, entire recycling rate, and GHG emission.

Traditional car-sharing services are based on a two-way scheme, wherein the user picks up and returns a vehicle at the same parking station. Some services also permit one-way trips; that is, the user is allowed to return the vehicle to another parking station. The one-way scheme is more widely available for users but may lead to an imbalance in the number of vehicles at each parking station; in other words, the user demand in each station may not be satisfied. In such cases, the service provider can relocate the vehicles to create a better distribution among the parking stations. In the case of electric vehicle (EV) car-sharing, such a problem is more complex because the travel range depends on the charge level of the battery. Moreover, EV relocation can lead to an imbalance in staff members between stations. Thus, staff members themselves need to be relocated between stations to perform vehicle relocations, considering the waiting time for charging the EV. In this study, we consider the charging and relocation problems of an EV car-sharing system. Additionally, we demonstrate the joint optimization of vehicle relocation and staff scheduling that can minimize the waiting time for staff and EVs at parking stations.

The COVID-19 virus has been rapidly spreading around the world, and the situation in clinical treatment sites is dynamically changing. In such a situation, the prediction and management of the required number of hospital beds have become issues in order to prevent the collapse of medical care due to the number of patients exceeding the capacity. On the other hand, cumulative flow analysis and on-demand flow management method, which is an extended management system, have been developed and used as a tool for visually managing the state of inventory in factories and warehouses. Onodera et al. (2020) adopted the on-demand cumulative-control method to the estimation of the COVID-19 patient population in a case study. However, the base number of vacant hospital beds is not predicted. This study models the number of COVID-19 hospital beds by using the on-demand cumulative-control method, and estimates required number of vacant beds from the viewpoint of inventory management. Firstly, the cumulative-control analysis is modeled by the input number of hospital beds as the supply, the output number of patients testing positive as the demand, and the number of vacant hospital beds as the number of flow which means the difference between cumulative input and output number. Next, using the on-demand cumulative-control method, the base number of vacant hospital beds is estimated to prevent the overwhelmed healthcare. Finally, a numerical experiment is conducted, and potentials for the appropriate and current management/policy are discussed based on the comparison between the required number of input beds estimated by the on-demand cumulative control method and the actual number of beds.

Effective methods of skill transfer are required at the manufacturing site of heavy industry with high-mix, low-volume manufacturing. Several previous studies have shown that giving unskilled worker simultaneous experiences in the skilled worker’s visual sense from a first-person perspective and force sense is effective for the skill transfer. This paper proposes a method of skill transfer by creating holographic moving images of workpieces and tools from the skilled worker’s motion. In order to analyze the effectiveness of this method, a control experiment by the assembly work of Tetris Block was carried out as a preliminary experiment. By using the proposed method and movie instruction which has already been put to practical use, each experimental participant’s assembly work was observed and compared, and the difference of two methods was analyzed from the three viewpoints of the work procedure, work time, and body motion. As a result, the effectiveness of the proposed method was confirmed from three perspectives; the margins of the body motion can be provided by the movement of holograms, the movement of holograms can communicate the skilled worker’s detailed motion information, and the skill transfer is possible in the virtual space environment which is not affected by real space. The future task is to verify the findings of this study by applying proposed method to actual field works.

As the world of artifacts becomes unstable, today the nature of artifacts and their ideal form are being questioned. The world of artifacts is comprised of human, material, money, and information (3 M&I) systems, and it consists of the balance between companies, households, the economy and its surroundings (the environment). This paper continues from the earlier publications since 2016 and proposes a DX-design method for the realization of a cockpit-type and demand-to-supply H = W robot based on artifact principles. In recent years, we have entered a second era, in which managerial enterprise bodies have come to operate like clockwork and in which resources are moved. Within this precise managerial integration, by looking at the nature of waste versus efficiency in clock systems, a dilemma (Nash’s zone) can be seen between knowledge system integration (waste, muda) versus sharing (efficiency). On the other hand, the modern industrial engineering (IE) of the method study that forms the crux of this time study is regarded to be mathematical programming (MP). Since the drifting management era, manufacturing has involved capital and labor, but in the newly manifest knowledge society, the means (methods) of production are shifting towards knowledge creation. Modes of management and integration are also made possible mainly through the workers who possess such knowledge. From the above, the humanized managerial enterprise body can be considered as being both cockpit-type and as a demand-to-supply corporate clock through the H = W-type embodiment of a humanized artifact. In this study, we propose a next-generation method and a new method of time study. Moreover, this new method is formulated using Matsui's matrix method of introduction-development-transformation-conclusion-BG-type story. The existence (how it is), and the purpose (what form it should take) of the exemplar artifact, which in this case is an enterprise robot, are clarified, forming a basis for future discussions, and which may provide a forum for debate going forward.

We are moving towards a future production in which machines are networked with each other and offer new possibilities for process control and optimization. These goals are achievable with OPC UA, which is currently increasing in popularity. To achieve this connectivity, interoperability between the machines must be guaranteed. OPC UA provides concepts for testing to achieve interoperability. However, these have not yet reached their full potential. In addition to being used for certification with the OPC Foundation, these concepts may also improve the development of OPC UA products in the future as well as the integration of such products in factory environments. In this paper, the tools and information in the domain of OPC UA Profiles and testing that are publicly available, are outlined and critically evaluated. They are put in context with use cases of OPC UA developers and standardization working groups. The findings are compared to examples with similar requirements in another domain, Linux package managers. This comparison outlines possible improvements in the handling of OPC UA Profiles and test cases, which are explained in greater detail.

Manufacturing systems are in the middle of a digital transformation. As systems in the assembly line are upgraded into cyber-physical systems (CPSs), capable of communicating between each other and carrying out complex computational tasks, the need for tight centralized control from an enterprise resource planning (ERP) and manufacturing execution system (MES) is less vital. In fact, not only manufacturing processes follow the trend toward decentralization and are moved to the edge layer. Other business processes along the supply chain have the potential to follow the digitalization process, such as procurement and supply flow management. This evolution brings new opportunities and challenges to the field. On the opportunity side, we identify shorter cycle times from product design to production, flexible production systems and multi-stakeholder production. Among the associated challenges, the collaboration of product, production, and business aware edge assets in multi-stakeholder environments stands out. This work proposes a new architecture for smart factories, in an environment where the products, supply stations and manufacturing equipment are controlled by different stakeholders. Requested manufacturing operations and supply flow are generated from machine-to-machine (M2M) negotiated business agreements between pairs of involved stakeholders. The manufacturing workflows are created and managed at each production workstation based on the smart product’s needs. Operations and supply flow progress is logged in distributed ledgers for the involved pairs of stakeholders, providing non-repudiation and immutable data on the M2M business agreement. The proposed architecture enables the automation of business processes providing benefits in terms of decreased transaction time and cost.

Manufacturing industry has been evolving during the last few centuries. Industry 1.0 started with mechanization and the use of steam power. Mass production using production lines and assembly lines dominated Industry 2.0 era. Industry 3.0 era brought automation, flexibility and product diversity and Flexible Manufacturing Systems (FMS) and cellular systems were extensively used. Recently, there is a shift towards the fourth industrial revolution (Industry 4.0). Industry 4.0 includes the combination of technologies working together to fulfill a manufacturing task. Industry 4.0 utilizes internet of things (IIoT), big data, cloud computing, cybersecurity, autonomous robotics, augmented reality, and additive manufacturing (AM). The purpose of Industry 4.0 is to integrate the entire network to function as one system. In this study, we are focusing on scheduling 3D printing machines, namely Markforged Mark Two printers. Process parameters that can be considered in these printers are layer height, infill density, print speed, build orientation, infill patterns, and print temperature. These machines are Fused Filament Fabrication (FFF) 3D printers. The parameters considered in this study are infill density and layer height. Infill density dictates the amount of material that is filled on the inside of an object while it prints. Infill density has a role in a part’s strength and weight. Generally speaking, the greater the infill density, the stronger and heavier an object will be. Lower infill densities on a part suggest that the object’s intentions are purely visual with higher infill densities meant for functional parts. Markforged Mark Two allows infill density for rectangular infill to be from 0–92%. On the other hand, layer height determines the amount of material that is extruded through the nozzle during each pass. Markforged Mark Two allows for three different layer heights to be examined, 100, 125 and 200 mm. Layer height plays a large role in print time as the amount of material extruded effects the completion rate of the object. Layer height’s impact can also be seen by a part’s fineness or detail. This is represented visually on the object by being able to see each pass of the plastic material. For example, an object with a larger layer height will look rougher and not as smooth as an object with a lower layer height. However, it is well known that a lower layer height increases print time whereas a larger layer height implies a faster print time. Several parts with different geometries and also sizes are included in the study. The scheduling performance measure considered is makespan. The objective of the study is to find the optimal parameter settings for multiple jobs such that makespan is minimized subject to minimum restrictions on print parameters for various jobs. A mathematical model is presented to minimize makespan first. Once the optimal makespan is found, the model is re-run such that better quality parameter settings are determined while keeping the optimal makespan unchanged. Later, the results of the experimentation with various parts are discussed and future work is recommended.

Recently the growing sensitivity of various governments toward a cleaner environment has increased the demand for electric vehicles (EVs). The battery is one of the most vital components in an EV. With the rapid development of EVs, the applications of lithium-ion battery (LIB) have become more and more extensive. Among the LIBs, the pouch type lithium-ion battery offers a simple, flexible, light weight, and robust solution to battery design, therefore it is considered to be the most promising technology for power battery. The pouch type LIB cell manufacturing processes include lithium battery cell assembly, electrolyte filling, formation, and aging, etc. The purpose of the formation is to form a stable SEI film on the electrode surface by charging. In the formation process, if the electrode tabs deviates from the accepted range, that will cause the failed charging. Therefore, a machine vision system should be built to automatically detect the electrode tab deflection.Recent developments in the field of deep learning have inspired a new interest in using neural network for general image classification tasks. In this paper, we adopt an instance segmentation algorithm called Mask R-CNN to detect and segment electrode tabs. Using the generated bounding boxes of the Mask R-CNN together with the Canny edge detection method, we can decide whether the electrode tab has deflected. As there was no existing dataset, we built a new dataset containing 398 pouch LIBs images. Among them, 200 samples were used as the training dataset, and the other 98 images were used as the validating dataset. Experimental results show 100% accuracy in electrode tab detection for both training and validating dataset. Another 100 pouch LIBs images were captured and used as test dataset. For the test dataset, the proposed algorithm has 100% accuracy in electrode deflection detection and 95% accuracy in polarity. It verifies the effectiveness of the proposed approach in electrode tab deflection detection for pouch LIB.

Numerous companies are thinking about using additive manufacturing in their production processes to leverage expected potentials such as higher flexibility, reduced costs around tools, jigs, and spare parts as well as a reduction of costs and times for manufacturing of complex goods. A substitution of an existing subtractive production process is what comes to their mind most of the time. They hope for a better material efficiency and an easier and cheaper way to produce small batches of products. The factual circumstances, however, are often much more complex than some companies, especially small and medium-sized enterprises (SME), initially assume. To give companies a clearer idea of the implications of using additive manufacturing, a rough filter model and a comparison of process chains of additive and conventional manufacturing is presented in this paper. The analysis consists of both the digital as well as the physical process chain and thereby provides a holistic picture of changes that are implied by the usage of additive manufacturing. In order to provide a clear reference for manufacturing companies, the report also discusses facts that have been discovered in cooperation with an SME during the production of real customer parts.

The current understanding is that automating inspection is one of the most important issues in the manufacturing industry, and many studies are being conducted for automated inspection. Recently, there has been a lot of research on detecting defective products from images using neural networks. Previous research has focused on detecting only two-dimensional defects (flaws, chips, etc.) from images and not three-dimensional defects (warpage, overlap, etc.) using images. Additionally, the equipment required for 3D inspection tends to be expensive, large-scale, and costly, so the initial cost is high. Many manufacturing sites have not been able to introduce it. In this study, we perform the 3D inspection using industrial cameras introduced in many manufacturing sites. Specifically, we will detect 3D defects such as war pages of transport trays used in actual manufacturing sites from 2D images. First, since the captured image contains noise due to many shooting environments, appropriate image processing is applied to remove the noise. The two types of images used in this process are those taken at the actual manufacturing site and those taken in a laboratory with a good shooting environment. We took the tray images in the laboratory at different angles between 10 and 90 in 10° increments. The processed images are then inputted to transition trained CNNs (Convolutional Neural Networks) for deep learning to perform binary classification of abnormal and normal. We also use Grad-Cam to visualize the learning to understand which part of the image the network focused on for classification. As a result, it shows that the network was unsuccessful in performing binary classification on any birds-eye view images. In comparison, the Grad-Cam visualization results show that the network obtained candidate features for tray warping, a 3D change, from the 30 to 40° images for the images taken at the manufacturing site and the images taken in the laboratory.

Constructing and optimizing an operation sequence is a major concern in production. However, designing high quality sequences is difficult and assembly sequence planning (ASP) is a NP-hard problem. Due to inherent process uncertainties production processes which include robotic manipulation of deformable objects, ASP can be even more complex than when only rigid objects are manipulated. Genetic algorithms (GA) are a commonly used heuristics for ASP. GA is suitable for ASP of robotic operations with deformable objects, which typically require addressing multiple objectives and which have complex production constraints. However, not all constraint may be explicitly known during the sequence design time. The current work examined ASP with different levels of known production process constraints. Integrating an arc consistency algorithm (AC3) for constraint satisfaction in the initial population generation process was implemented and compared to random population generation. Integrating process duration and the longest common sub-sequence (LCS) index (to sequences of similar products) in the fitness function was implemented and different integration methods were compared. Results show that the effects of using AC3 for initial population generation depend on chromosome length and are not related to the rate of addressed constraints. For long chromosomes AC3 based generation is considerably faster than random generation. The impact of adding LCS to the fitness function depends on the rate of constraints addressed and is not related to chromosome length. When not all the production constrains are addressed LCS increases the number of feasible solutions obtained and is not related to chromosome length.

In the case that abnormality in a state of systems is found, identifying when the state in the system changed, that is, identifying the change-point is helpful to investigating the cause of the system abnormality and dealing with it. In our research, a method of identifying change-points and tracing the system state has been discussed based on the data from the system, when the characteristic values representing the system state obey the Poisson distribution. Specifically, based on the likelihood theory and information statistics, a method of selecting the best statistical model for explaining the state changes of the system has been proposed as the state transition tracing method. Then, the effectiveness of the proposed state tracing method has been verified through numerical simulation. In addition, an application example of inferring the cause of state changes from the outcome of our proposed method has been shown.

McKone et al. (J Oper Manag 19:39–58, 2001) proposed Total Preventive maintenance (TPM), Just in time (JIT) and Total quality management (TQM) to contribute significantly to manufacturing performance (MP) and TPM could be considered as a part of the manufacturing strategy. The use of preventive maintenance in equipment maintenance could effectively reduce machine occurrence and reduce machine efficiency due to failure (Niu et al. in Reliab Eng Syst Saf 95:786–796, 2010; Panagiotidou and Tagaras in Eur J Oper Res 180:329–353, 2007; Swanson in Int J Prod Econ 70:237–244, 2001). Many studies in the past have applied the concept of total preventive maintenance (TPM) to equipment maintenance to reduce downtime and improve machine efficiency effectively (Panagiotidou and Tagaras in Eur J Oper Res 180:329–353, 2007; Swanson in Int J Prod Econ 70:237–244, 2001). Utilizing preventive maintenance can reduce machine’s shutdown and improve the equipment efficiency. The traditional total preventive maintenance methods focused on maintaining single component. The research, however, strives to maintain a group of components to further reduce the maintenance time. The components were clustered into group according to their distributions of lifespans. The clusters that saved the most maintenance costs are recommended to managers for maintenance scheduling. The methodology was applied to an auto component company for experiments. The results showed that OEE was improved from 81 to 84%.

This paper presents the process simulation and flow of material in the cavity of the mold during the compression molding process. The composite material used as the charge in the molding process is comprised of recycled polymers waste extracted from various automotive parts and reinforced with engineered wood fibers. The compression molding tool is a re-design version of an automotive battery cover that is currently in application in various car models. The newly designed component is suitable to be manufactured using recycled polymer composite material. Some additives were added to the composite material to make it more viscous and to aid in the flow of material in the mold. The simulation models are built-in Moldflow with the same part geometry and processing conditions were kept the same for all the composite blends. Later, all the composite blend models were compared with only recycled polymer models to see the effects of fiber material on the output of the compression molding process. The final simulation shows that the 10% fiber content in the composite material exhibits the most promising option with no voids or cavities and less fill time of the entire part geometry.

The in-line stocker is a new type of the automated material handling system that has begun to be used in the display industry. This not only moves unit loads under processing across different locations within manufacturing facilities, similar to the automated guided vehicle (AGV), but also stores them on its shelves, similar to the automated storage and retrieval system (ASRS), which can significantly reduce space for material handling. However, traffic rates are generally very high inside the in-line stocker and two robots serve along the single lane, which is the dual-robot in-line stocker (DRIS). One difficulty in applying the DRIS to shop floors is that the exact transport capacity of the unit DRIS is not known. This paper develops an analytical model to estimate the capacity of the DRIS based on a combined scheduling and simulation method. It calculates movements of two robots over time in the space consisting of time and location and precisely measure necessary time for waiting or backtracking to avoid collision of two robots. An experimental analysis was conducted to validate the correctness and usefulness of the model based on the data used in an actual manufacturing site. The analysis result illustrates that the average processing capacity of the DRIS increases compared to the SRIS (single robot in-line stocker) as the length of the DIRS increases, which is consistent with the expectation of practitioners in the industry. The paper also verifies that it is necessary to carefully determine the operating specifications in actual uses since the transport capacity of the DIRS varies considerably by its operating parameters, which can be optimized by the analytical model.

It is possible that workers’ error, variation of processing time, lack of parts and machine failure will affect the delivery date of each production process. The consecutive delay in the process can lead to the postponing of manufacturing production. By optimizing the workers’ assignments, it is possible to achieve the delivery date of the product and reduce the total expected cost of the product. This study takes a parallel production line as an example and considers the problem of how to optimally assign workers to each process when the total expected cost is minimized by having the same number of processes in each production line.

The inventory model for deteriorating items with a stochastic demand rate is studied in our paper. Inventory models dealing with the deterioration of items have attracted considerable interests in recent decades since the deteriorating phenomenon is a crucial factor affecting the profits of a company. However, most of current inventory models considering deterioration assumes a certain or a constant demand function which is unreasonable in the prevailing market that conforms to our daily reality. Therefore, stochastic demand must be considered. Besides, shortage is allowed in the study. We provide approximating solutions for optimal ordering quantities for our model, and show that our optimal solutions from the assumed models give conditions and results very close to the optimal solutions obtained by computation. Further, these results illustrate the impact of various parameters on the optimal policy and the profit.

Due to its high potential to perform many tasks faster, more accurately and in greater detail than humans artificial intelligence (AI) has been attracting growing attention across industries. In manufacturing, AI, in combination with digital sensors, networks and software-based automation, defines a new industrialization age. The integration of AI into production processes promises to boost the productivity, efficiency, as well as the automation of processes. However, AI adoption in manufacturing is currently still in its early stage and lacks practical experiences. This raises the question, to which extent manufacturing companies are ready to implement AI. While approaches to assessing the maturity in terms of the digitalization or Industry 4.0 (I4.0) of manufacturing companies are well established and discussed in the literature, approaches that specifically address AI in manufacturing are still lacking. To address this gap, we present an approach to analyze and monitor the readiness of manufacturing firms for working with AI technologies. In accordance with the existing assessment concepts of digitalization and I4.0, our approach examines different areas of digital technologies on the product and production level of manufacturing firms. Moreover, it incorporates the key foundation for AI—security and data—into a conceptual model. We generally assume that companies need to achieve a certain level of digital readiness in three key dimensions in order to be ready for implementing AI-based technologies. We operationalize these dimensions through a variety of product- and production-specific as well as data- and safety-related indicators. In order to illustrate the implementation of our concept in practical terms, we present the results of the readiness assessment of two German manufacturing companies.

The evolution to Industry 4.0 is creating the impetus for the manufacturing industry to increase productivity through smart management and stabilization of resources, capacity and utilisation. Increased plant availability, extended service life of resources as well as optimised product and process quality require intelligent maintenance strategies. The conventional reactive maintenance (run-to-failure) causes unexpected production stoppages, and preventive maintenance at times leads to waste of working hours and material due to the premature replacement of machine components. A smart Predictive Maintenance (PdM) strategy equipped with fault detection and prediction based on acquired, processed and analysed data can result in an accurate estimation of the Remaining Useful Life (RUL) of machine components and thus trigger appropriate maintenance action plans. Data acquisition, processing, analysis and rule-based decision supporting require the development, application and combination of various Industrial Internet of Things (IIoT) devices, models and methods in an integrated manner. Through transparent development and integrated harmonisation of all models, methods and technologies, fault detections and respective RUL estimations of machine components become more accurate and reliable. This leads to an increasing acceptance of employees towards software-based recommendations, in particular maintenance instructions for operators and proposals for an optimised development of the next generation of production systems and equipment. Within the scope of the EU-funded project Z-BRE4K, this paper proposes an IIoT architecture that presents models, methods and technologies in an integrated manner and highlights the data and information flow between them. The architecture including the infrastructure has been applied in three pilot cases with the industrial end users PHILIPS, GESTAMP and CDS to demonstrate the compatibility of the architecture to different industries with various production systems and diverse conditions, requirements and needs. Based on the adaption of the generic architecture for the pilot cases, the models, methods and technologies were developed efficiently and continuously improved and validated. The proposed architecture is intended to be applicable across industries to facilitate the transformation from reactive or preventive to PdM and thereby improve the competitiveness of manufacturing companies.

The cumulating carbon fiber-reinforced polymer (CFRP) composite wastes need to be disposed efficiently. So far, the most effective thermal-based recycling technique, namely pyrolysis, has grown exponentially towards industrial scaling in developed countries such as the UK and Germany. Typically, even the slightest mistakes can cause unfavorable results and delays in workflow within such a massive operating environment (e.g., >1 tonnes/day operating capacity). The existing semi-automated and, in some cases, fully automated plants should be continuously updated to resemble the varying classes and volume of the CFRP composite wastes. To overcome such research gaps and imprecise manual errors, Internet-of-Things (IoT) based framework is proposed. This paper studies the theoretical implementation of an IoT-based framework into the pyrolysis process to recycle CFRP composite waste to manage the process based on the principles of cyber-physical systems. The proposed framework consists of sensors and actuators that will be used to collect the data and communicate with a central management constructed as a platform that will articulate and manipulate data to satisfy the requirements of the recycling process, computationally modeled through logical relations between physical entities. In this case, the management unit can be either controllable or monitored remotely to increase the operation time of the plant. Our objective is to propose a scalable method to improve the recycling process, which will also help future decision-making in handling recycled carbon fiber. Specifically, this study will go beyond the state-of-the-art in the field by (i) automatically calculate the mass of the waste and adjust the operating time, temperature, atmospheric pressure, and inert gas flow (if needed), (ii) regenerating heat so that after the first batch is recycled, the resin high in calorific value will be burned and will be releasing energy, whose generated heat needs to be trapped inside the furnace and then regenerated into the system, and (iii) decrease energy consumption and fasten the process flow time. In summary, the proposed framework aims to provide a user-friendly control and temperature monitoring that can increase the overall efficiency of the process and avoid possible process shut down or even char formation by a controlled atmosphere in the pyrolytic reactor.

For achieving the ambitious objectives of intelligent production, the artificial intelligence through its machine learning algorithms represents one of the most promising technology. The employment of machine learning algorithms for the optimization of complex production processes faces the big challenge of selecting the suitable machine learning method which fits the optimization parameter objectives. This paper introduces our approach for automated selection of ML algorithms to be used for optimization of a specific production process. The approach and the corresponding method consists of the following main blocks: (a) production process definition, (b) ML performance, (c) selector constructor and (d) assessment and incremental improvement of selector performance. The first component defines a typical production process or domain based on a well-established set of features, e.g. product quality inspection through process features as accuracy, material characteristics, etc. The ML performance construct contains precise defined performance of well clustered ML algorithms based on established benchmarking. The third construct, the Selector, automatically realizes a perfect mapping between the production process features and the performance of the ML algorithms. The logic of this automated selection represents the innovation of our work. The last component assesses the selector algorithm based on a set of specific KPIs for each production domain or process. The incremental improvement of the selector is approached as well, closing the loop between all method components. The developed approach and method have as foundations our work on identifying critical production processes/domains as core of realizing the intelligent production and laborious developed collection of ML algorithms, based on their performance data. These foundations and a motivation scenario are presented inside our paper to highlight the relevance of our research work.

Competitiveness of an electronics manufacturing services (EMS) firm is being closely monitored and interrelated in production cost control and product reliability performance level. Data analytics in massive manufacturing data can extricate huge business opportunities and values to the firms. Major challenge in data analytics application is heterogeneous and enormous of which dynamic data generated from continuously running production line reflects real-time velocity of production environment, so that the factory management demands data analytics to provide real-time solution for the improvement right on the spot. Cloud-based data analytics exhibits problems such as data capturing, storage, transfer latency and data quality that hinders the advancement of big data analytics in smart manufacturing horizon. Selection of appropriate data mining algorithm or techniques has been challenging to industry leaders in deriving desired patterns or model solving the exact problem they are facing. The aim of this research study is to illustrate the edge-based intelligent integrated information framework (INFO-I2) for the improvement of data quality in relevancy and enabling cloud-based data analytics to focus on product performance augmentation (Pipino et al. in Commun. ACM 45:211–218, 2002 [1]). In the case study, edge devices had been used for not only real-time data collection but also localized failure analysis and predictive analytics to perform autonomous decision-making in different workstations through production process. Cloud-based computing performs efficient optimization analytics for product functionality performance with those processed data which is an integrated production management system and knowledge base generated from localized data analytics of edge devices. The implementation of edge-based information framework improves the workflow management, eventually reduces manufacturing cost and improved product reliability. The contribution of this paper is to demonstrate how the proposed cloud-based manufacturing system architecture adopted both Cloud and Edge Computing to enhance product reliability and pave the way for smart manufacturing.

Current semiconductor shortages, trade obstructions, as well as the COVID-19 pandemic have shown the urgent requirement to rethink existing economic and production structures as well as trade practices. Such global events revealed the susceptibility of value-added structures to environmental disruptions. Especially the complex and highly optimized value chains in the automotive industry were particularly affected. Lack of early warning systems, missing redundancies and ever-increasing dependencies between individual value-added partners can be seen as some of the reasons for the enormous effects the industry is facing. However, looking into the future, fast and flexible product changes might affect the supply chain in similar ways as the current events. To secure value creation and employment and to minimize the impact of future crisis, the industries structures must be made more resilient. This paper provides a contribution to this through an integrated approach. This approach combines elements of linear value creation with a lifecycle-oriented perspective: The life cycle perspective considers temporal changes and dependencies between the individual supply chains of the life cycle phases. Subsequently, the value creation system is decomposed into manageable elements. The corresponding dependencies over selected life cycle phases are described. By creating scenario-robust value-added modules, risk within value-added systems is minimized. To illustrate the approach, a case study on automotive software and electronics is presented. Future changes regarding possible functions of an electric vehicle are anticipated and technical implications are derived on basis of a scenario analysis. Subsequently, the value-added module of integrated software and hardware development are examined in more detail and possible value-added configurations are displayed to indicate that the developed approach is suitable to generate robust supply chains. Interdependencies between value chains of multiple life cycle phases are discovered. Consequently, our approach provides additional value in the context of flexible products and services as well as their associated value-added processes and systems.

This article investigates theory of agency as a theoretical underpinning in the field of supply chain finance. Specifically, through a hermeneutic approach, the authors examine the development of the theory, its postulations and assumptions, and its use in the field of supply chain finance. This leads to three conclusions: the theory is currently adopted rather superficially; there is potential for further developments by investigating non-standard configurations; and there is an increasingly relevant set of articles in supply chain finance that seems to be positioned within the boundaries of the theory of agency but makes no mention of it.

The world experienced historic challenges in global demand management as a result of the covid-19 pandemic. Supply chains were abruptly interrupted with new traffic rules among countries and organizations confronted hard moments of absolute uncertainty with an extreme complex planning scenario while customers demanded resources on time in order to stay safe at home during confinement. For this reason, supply chain risk management and demand forecasting with artificial intelligence has become even more explored by the scientific community. In this context, this paper proposes an investigation of machine-learning projects’ contribution for supply chain management in organizations, not only during the pandemic crisis, but during the last recent years. PRISMA approach was applied for a systematic literature review, limiting English written articles indexed at Scopus and complementary sources, such as Science Direct and IEEE. Results have defended the increasingly important role of machine-learning projects in supporting organizations to plan their operational demands and activities, improving operational efficiency and strengthening strategic supplier selection even in challenging pandemic times. The main contribution is focused on examining theoretical relationships among recent approaches and address mutual strategic achievements through a diagram. Results presented by a summarized diagram exposed machine learning strategic value for demand forecasting, supply risk mitigation, lead-time reduction, greener operations and strategic supplier selection. Some common best practices observed revealed training and test segmentation, feature importance analysis and dimensionality reduction. Limitations are linked to further research suggestions, increasing the present bibliographic portfolio selection along with case-study implementations in order to extend connections between theory and practice.

Eucalyptus is the raw material for sectors such as furniture and cellulose, but it is also an important input for thermal energy, being used in several other industrial segments. The high price of agricultural commodities and the low price of eucalyptus affect the renewal of forest areas. The risk of supply in Paraná (Brazil), makes the cereal producers located in the state seek to implement strategies to ensure the supply of eucalyptus chips for drying soybeans and corn. The present work proposes a mathematical model for optimizing the allocation of forest units, reducing costs of leasing, and transporting chips between forest stands and grain receiving units in an 8-year production cycle, considering cuts in forests with ages ranging from 5 to 8 years. The model was validated using a case study of a company located in the Paraná state, in which the optimization guaranteed the supply of chips in 10-grain receiving units and resulted in the choice of 7 out of 14 forest units available for leasing. The model also allows the analysis of the company’s business policies, such as verifying the impact of changing the minimum area for harvesting, or minimum transportation lot between a point of origin and a single destination.

In recent years, the environment has become a hot topic among the government, society, and enterprises due to global warming, ozone depletion, and air pollution. Businesses are believed to be at the root of most environmental problems. To reduce environmental pollution and promote sustainable development, more and more TV manufacturers are adopting a green supply chain to achieve sustainable development and produce more environmentally friendly television (TV) sets. However, some green TV manufacturers are underfunded in the development and production process. In this paper, a green TV supply chain system composed of a financially constrained manufacturer and a reputable retailer is proposed to help the TV manufacturer successfully produce more green TVs. The supply chain model is established by using two financing tools. A time-dependent residual value is considered in the model, and the later the clearance time, the lower the residual value. We investigate the effects of time-dependent residual value on operational and financing decisions and measure the supply chain system’s profit risks. The following results are concluded: (1) The clearance time of unsold TV sets affects the retailer’s and the manufacturer’s decisions. (2) The financing equilibrium can be obtained under certain conditions. (3) The profit risks of the retailer increase with the order quantity and the clearance time. Numerical analysis is used to verify the derived results.

This paper discusses a distribution planning model with n distribution centers (DCs) and m customers. Ideally, each DC serves customers in its zone. Each customer order may include multiple items. However, as the need arises, customers can also be served from other DCs to avoid delays to the customers. Besides, items in a customer’s order may also come from multiple DCs not to violate safety stock values established for each item in each warehouse. Furthermore, there are alternative transportation modes available with different costs and CO2 emission. Environmental issues have become increasingly popular and various studies have been proposed in the literature in recent years. Many corporates are increasing their sustainability efforts and becoming more responsible members of the society. We assume stochastic transportation leadtimes. They are assumed to be normally distributed with known mean and standard deviation values. There are multiple objectives considered in this study and they are: (1) minimize total shipping cost, (2) minimize probability of tardiness, (3) minimize the environmental impact of transportation, (4) minimize safety stock violations. A multi-objective fuzzy mixed-integer-non-linear programming model is developed to handle all these objectives such that satisfaction level is maximized. The task is to determine which DC(s) will serve each customer. Two experiments are conducted, (i) the same network is experimented with multiple periods where inventory levels of items are revised at the end of each period based on the shipment made, (ii) multiple experiments are conducted with various parameters including number of customers, number of items in orders, transportation leadtime distributions, starting inventory levels, safety stock values and fuzzy operators. One of the important features of this study is that it combines both stochastic and fuzzy parameters in the same mathematical model, which has been rarely found in the literature.

Since the Great East Japan Earthquake in 2011, the importance of BCP has been recognized widely, as a result, formulation rate of BCP in Japanese manufacturing industry has increased to about 50%. However, nearly 80% of them are major manufacturers. There is a difficult task on supporting small and medium-sized manufacturers in developing BCP. In addition, in recent years, supply chains have become much more complex increasingly due to the globalization. Not only large-scale supply chain disruptions such as natural disasters or industrial accidents, but also local and small-scale disruptions such as power outages or information system failures can cause enormous losses. Even if a BCP is formulated, whether it is an appropriate BCP or not has not been sufficiently discussed quantitatively. To support the manufacturers to formulate an appropriate BCP, in this study, a stochastic model of supply chain disruptions and a BCP model are created based on a supply chain consists of component suppliers, retailers and an electronics device manufacturer as the main participant. Loss mitigation effects by different levels of BCP are analyzed to help the manufacturer to formulate appropriate BCP according to the attributes by numerical simulation.

There is a lot of evidence of the manufacturing strategy shift from mass production to the individual and decentralized micro-production, which in essence means a shift from low-cost country production to production at the place of consumption. This paper presents a vision for a highly automated digital urban manufacturing network, based on a central market mechanism for matching designers, customers, and producers operating digital-manufacturing machinery. The mechanism proposed automatically clears the market by matching supply and demand, based on pre-given or auction-based bids with an optimal allocation. Contracting, profit-sharing and settlement are automated according to pre-specified market rules. Network participant profiles are used and updated to establish reputation and trust that are used in the allocation optimization. This paper presents the elements and logic of an automated digital urban manufacturing network on a technical level and discusses some of the benefits reached by shifting centralized production to decentralized urban networks in a wider perspective. Practical implications to the real-world industries, where the presented vision is relevant are shortly discussed.

Living labs are not new. However, they are often framed in literature as methods for user-centric innovation management. We argue that a living lab can be seen as a research methodology within the field of cooperative inquiry, as an operationalisation of concepts such as action research and engaged scholarship. To do so, we present the case of a living lab in the Netherlands, focused on data sharing technologies, and analyse how it generates research output with clear pragmatic value as well as addressing interesting research challenges. Finally, the paper positions the methodology against existing concepts within cooperative inquiry.

Firefly algorithm (Firefly Algorithm; FA) is one of the swarm intelligence algorithms. FA is known as the algorithm that is based on the flashing behavior of fireflies and their associated behavioral patterns. Specifically, FA associates the objective function with the light intensity of fireflies, and searches for the optimal solution based on the feature that fireflies are attracted to those with higher light intensity than themselves. In this way, we can find the optimal solution. However, since FA only uses the information of the other firefly with light intensity stronger than itself, it may not be able to find the optimal solution if the firefly with the highest light intensity falls into the local solution. On the other hand, while fireflies in nature are naturally distinguished into males and females, in general FA, all fireflies are defined as the same functional group with no gender and uniform behavioral patterns. Therefore, FA do not have the diversity of behaviors and may not be possible to properly solve problems due to uniform behavior patterns. In this study, we propose an algorithm that distinguishes the sexes of fireflies in FA and defines different behavioral patterns for female fireflies. By achieving diversity in the movements of fireflies, we aim to avoid the defined FA system falling into local solution and find the optimal solution. The proposed FA system is applied to the maximization problem of mathematical function, and the effectiveness of the system is verified by comparing it with existing FA systems in previous studies.

Overall Equipment Effectiveness (OEE) is a performance evaluation tool that measures equipment productivity with key indicators. In addition to construction industry, OEE has been widely used in different industries, such as manufacturing, food processing, semiconductor, medical service, etc. However, the characteristics of construction industry are the same as that of most manufacturing, that is, profits and revenue come from the effectiveness of machinery and equipment usage. Therefore, this research develops a method called equipment performance management system (EPMS) to evaluate the performance of the propulsion equipment of pipe jacking tunneling using OEE. This EPMS is applied to a real world case study to demonstrate how to define, calculate, and analyze these performance indicators of the pipe jacking tunneling machinery and equipment. The case study proves the feasibility of the research method, which is helpful for reflecting operating performance and identifying root causes of potential problems in time so that immediate improvement or rectification actions can be implemented to avoid further losses of the company.

Besides the sequence itself also additional factors serving as moderator variables affect the value of scheduling objectives. For mixed-model assembly lines, especially number and heterogeneity of different products, their volume mix proportions, average workload of the jobs to process and the degree of grouping of identical jobs within the sequence play a major role. By means of a simulation study based on data from a real unpaced mixed-model assembly line in the automotive industry, this work analyzes the impact of these moderating variables on various scheduling objectives. The analyzed scheduling objectives encompass flow-related objectives like mean flow time, productivity-related objectives like makespan, customer-related objectives like mean earliness, the supplier-related objective part usage rate variation and the human-related objectives mean learning effect and mean deterioration effect per job. Simulation scenarios are defined that differ regarding number and heterogeneity of products from three homogeneous to seven more heterogeneous products. Within every simulation scenario the volume mix proportions of the products, and inherently also the average workload of jobs, are systematically varied. Every simulation scenario is analyzed for five sequence types differing in the degree of grouping of identical jobs. For almost all scheduling objectives, strong dependencies on the volume mix proportions can be perceived, particularly for mean flow time. Homogeneous volume mixes with a dominating product in the mix often lead to other objective values compared to heterogeneous volume mixes that allow using alternation effects between various products in a sequence. Concerning degree of grouping, while some scheduling objectives like part usage rate variation are always strongly affected by the degree of grouping for every volume mix, other objectives like throughput show strong dependence only for some mixes and makespan does not even show any tendency. Average workload plays a less important but still recognizable role as moderator variable for most objectives except for throughput for which workload is a major explanatory factor. Number and heterogeneity of different products has a strong impact on mean learning and deterioration effects.

Recently, a dual linear consecutive system with three failure modes was proposed. This system contains two parallel subsystems consisting of $$n$$ components arranged in a line. The system fails if (1) subsystem 1 has at least $$k_{1}$$ consecutive failed components, (2) subsystem 2 has at least $$k_{2}$$ consecutive failed components, or (3) the system has $$m$$ consecutive pairs of failed components. This system has diverse applications, such as road lights on highways. To design a reliable system, this study addresses the problem of finding the component arrangement that maximizes the system reliability, assuming that the components are functionally exchangeable. A simple enumeration method can theoretically find the optimal arrangement, but it is computationally expensive and impracticable for larger numbers of components. Furthermore, no algorithm for efficiently finding the optimal arrangement of a dual linear consecutive system with three failure modes has been established. Thus, this study develops a branch-and-bound-based algorithm to solve the component assignment problem for a dual linear consecutive system with three failure modes. For an efficient reliability computation in the proposed algorithm, we combine a recursive equation with the branch-and-bound method. This results in the elimination of redundant computations. Furthermore, we derive a condition for pruning based on system reliability, which can reduce the number of enumerated arrangements. Through numerical experiments, we confirm the usefulness of eliminating redundant computations and pruning based on system reliability by comparing the proposed algorithm with the enumeration method. The proposed algorithm can improve the reliability of real-world systems expressed as dual linear consecutive systems with three failure modes.

Hazardous Waste Management is defined as the safe and efficient handling of hazardous waste to decrease its toxicity to humans and the environment by means of proper transportation, processing, and disposal. Hazardous waste is generated by different sectors, including the medical sector, industrial sector, domestic sector, among others. Industrial hazardous waste consists of chemicals and compounds of a complex structure, which poses a great risk for public health and the environment alike. This potential risk heavily influences decision-makers when choosing a suitable location for establishing a waste processing facility. Therefore, the process of designing a hazardous waste transportation network comes with many challenges. This paper presents a profit-oriented mixed-integer linear programming model for the hazardous waste location-routing problem, with the main objective of maximizing the overall profit in the network and conditions focusing on minimizing the associated risk in terms of population exposure. The transportation network includes waste generators which are, in this case, factories, along with three different types of hazardous waste processing centers: treatment, recycling, and disposal centers. The formulated problem is coded in Python and optimally solved by Gurobi Optimizer. Furthermore, to deal with the NP-hard nature of the problem for large numerical instances, a metaheuristics algorithm based on non-dominated sorting genetic algorithm II (NSGA-II) is applied, and the model is solved with both NSGA-II and Gurobi to investigate the improvements done by utilizing the genetic algorithm. A case study is conducted for the textile industry in Jordan, as to put the proposed model into practice.

In as early as the 1980s, air traffic flow management actions (ATFM), as supplementary strategies to match the demand for air travel with the available resource capacities, have been widely discussed and evaluated based on its implementation and probable trade-offs between conflicting and diverse interests of stakeholders in the commercial aviation industry. Among the ATFM actions—ground holding, airborne holding, speed controlling, and rerouting—rerouting is found to be a viable recourse particularly when flights are already at its en-route phase, where the presumed and more favored based on safety considerations, holding of flights on the ground, becomes completely infeasible. Some research works put forward relevant solution approaches including deterministic and stochastic mathematical programming models, machine learning algorithms, and simulation models. Despite the relevance and validity demonstrated by such models in testbed environments, even on a large-scale basis, these models failed to sufficiently capture the individual and collective interests of stakeholders altogether. Considering that the decision process in the air transportation system is taken part by stakeholders (i.e., airlines, air traffic control), previous research works tend to satisfy only one stakeholder by incorporating one or more of its interests (e.g., cost minimization, reduction of distance traveled). Such a case does not take full regard to how a stakeholder-specific solution might affect another stakeholder’s preference. Therefore, this paper aims to address the post-departure aircraft rerouting problem by proposing a multiple stakeholder-based target-oriented robust-optimization (MS-TORO) approach that incorporates the individual interests of stakeholders. A hypothetical case study is conducted to illustrate the proposed model. It can be noted that a significant shift of route preference occurs as goals are aligned in terms of the individual interests of the stakeholders and that of their collective goal. The results of this work can provide practical insights to stakeholders in the course of decision-making in a particular area of the air transportation domain.

The present paper describes a method for forecasting the economic number of times a sheet metal, integrated in a structure, and subjected to high corrosion, must be repainted before its thickness reaches a minimum limit, advising for its replacement. Within this approach, one is acknowledged of all the relevant costs originated, predictably, over the life cycle of the sheet metal, demonstrating the utility of combining descriptive functions (of the physical phenomena from the field of materials engineering) with financial analysis’ functions, thus completing what one can properly call—technical–economic analysis—whose opportunity is verified so many times in engineering economy and maintenance.

The advent of industry 4.0 facilitates the use of collaborative robots near humans. But this proximity brings about some critical challenges. The task allocation problem is one of the top challenges that deserve the serious attention of researchers to attain a successful human–robot collaboration. The main aim of this study is to summarize the existing literature on the topic of the task allocation problem in the human–robot collaboration context. In this paper, we reviewed the task allocation problem from the perspectives of definitions and terminologies, different types of allocation methods, evaluation criteria, implementation procedures, and application phases.

Drones are recently used for monitoring incidents in smart cities such as crime, transportation, disaster, etc. In this study, we present a monitoring problem by using drones. The monitoring problem is represented as a submodular maximization problem with a partition matroid constraint. The problem is known as NP-hard in general. If incidents that come up in a city lead to severe consequences, more effective solutions to monitor the incidents are needed to prevent as much damage as possible from the incidents. We develop an exact algorithm to solve the monitoring problem. We present two types of valid inequalities that are used to construct the exact algorithm. We also prove that the two types of the valid inequalities are valid in the monitoring problem. By using (near)-optimal solutions of drone operations from the exact algorithm, we can minimize losses from the occurrences of incidents. Numerical tests are also presented to show the performance of the algorithm.

Cross-efficiency (CE) evaluation is an extension of data envelopment analysis (DEA) used to fully rank decision-making units (DMUs). The ranking process is carried out on the matrix of CE scores, where an ultimate efficiency score is computed for each DMU through an adequate amalgamation process. In this paper, we propose a ranking procedure that computes the ultimate scores over two aggregation levels, which involve ordered weighted averaging (OWA) induced by different preference settings. First, the preference voting system embedded under the CE matrix is employed to derive aggregate votes as inner quantifiers of the consensual importance of the DMUs. Next, the preference order induced from the aggregate votes is adopted as a ground for rearranging the rows of the original CE matrix prior to the ultimate calculation of the efficiency scores. To substantiate the impact of subjectivity on the structure of the ranking patterns and, hence, assess the robustness of the proposed procedure, the induced OWA is implemented with different optimism level values. The proposed methodology is applied for selecting the best material handling equipments (MHE) within a sample of 25 real-life MHEs, collected from catalogues of MHE manufacturers and vendors.

In this study, the inventory management problem of a wholesaler company that supplies and distributes products according to uncertain demand is discussed. This study constructs a multi-product multi-period (s, S) inventory policy considering items with uncertain demand and determines the maximum inventory level (S) and reorder point (s) for each item to minimize the total inventory cost for the wholesaler company in question, considering budget constraint. (s, S) policy means that an order is triggered as soon as the inventory position declines to or below the reorder point (s) and difficult to determine the (S) values. The order size is chosen so that the inventory position increases to (S). Since it is a multi-product and multi-period system, and demand has stochastic structure the results getting from mathematical model are validated by using simulation optimization model.

This article presents a real-world case study for a multi-manned assembly problem issued in an engine manufacturing company: the balancing optimization for a mixed-model production line. In order to describe the real line-balancing problem, the article focuses firstly on the presentation of business needs, and layout and model restrictions assigned. A specific feature called dual station synergy is identified during an early phase of development, which requested mathematical model tailoring, herein described. The use of a mixed-integer linear programming model to solve the problem is presented, as well as its satisfactory results, with reduction of 9% in the cycle time with respect to the current practical solution. As a highlight, this study relied on the physical validation of the proposed mathematical solution in a productive environment of diesel engines.

Agile Project Management (APM) is a standard approach in the software industry, and because of its success in allowing projects to better meet its requirements, its application has lately been extended to other industries, without prejudice of its underlying principles and advantages. Among APM practices there are: iterative and incremental deploying the product, presence of informal communication and low bureaucracy and use of visual tools. Additionally, sprints allow the exercise of creativity to solve the presented problem and just-in-time planning to better use people’s time. However, adapting practices that were designed for developing a software to other types of products poses many challenges. To shed some light as to how agile practices may be adapted to different contexts, this work conducts a case study of a project in manufacturing that adopted an agile approach. The objective is to analyze how agile practices were applied in this project, from the factory floor to the car dealership. The analyzed project aimed to improve the quality of the after sales services of the company, reducing the percentage of customers problems and complaints related to the service. A semi-structured interview was conducted with the agile coach responsible for the project, which divided the analysis into five areas of study: work and efforts, materiality, agency and creativity, knowledge, and interests and power. Results showed that the highlighted areas are interrelated, since their benefits intertwine while achieving project success. These observed benefits match the application of APM in a software project and were related to practices such as the use of small teams, periodic short meetings, and tools for informal communication. These characteristics made improved project agility and eliminated unnecessary formal documentation and communication. Nevertheless, not all the characteristic tools of APM were used, for example, there was a lack of a project’s backlog and the utilization of an agile board, even though visual monitoring was used during the project’s execution. Future research is proposed to amplify the study, analyzing other types of projects, and develop a framework with the best practices found to implement the APM methodology outside of software projects.

This paper introduces a new form of bundling. Our approach is dynamic and contrasts with the conventional notion that bundling should be static over an infinite selling season: One may operationalize two major product line strategies partial mixed-bundling and mixed-bundling dynamically by considering their interaction with firm’s inventory information. Our contributions are as follows. First, we develop a model to describe operational bundling and show how one can reduce the complexity of the optimisation problem to obtain joint optimal inventory and bundling strategies. Second, we compute the stationary probability of the joint inventory levels and develop the long-run average payoff rate. Third, we provide necessary and sufficient condition for optimality of operational bundling, which becomes keystone to access the value of static is bundling. Under some conditions, our work shows that bundling strategy should always be dynamic when inventory costs are taken into account.

In this paper, we model and solve a multi-objective personnel planning problem in a military context to find appropriate values of recruitment and promotion policies that simultaneously minimize personnel cost and asset unavailability (due to the personnel shortage to crew assets). We propose a hybrid solution approach that combines a simulation–optimization method with data envelopment analysis (DEA). The simulation–optimization method integrates a genetic algorithm (GA) with a system dynamics (SD) simulation model. While GA can effectively search very large decision spaces, the SD simulation simulates the personnel system and its connections with the fleet of assets. In an interactive process, GA generates solutions of recruitment and promotion policies and sends them to the SD simulation, where the values of personnel cost and asset unavailability are calculated for each solution. Then, DEA computes the efficiency scores of solutions using recruitment and promotion values as inputs, while personnel cost and asset unavailability values as outputs. These efficiency scores are used as fitness values to guide the search process in GA. We test the proposed model on an illustrative example. The numerical results indicate the applicability of the proposed method in identifying the efficient solutions and reducing personnel cost and asset unavailability.

Modern production technologies, based on Industry 4.0, allow to increase the flexibility and resilience of production systems. These enhanced capabilities, along with a greater digitization of the shop-floor, give significant leverage to massive personalization of production. This work addresses an extreme personalization problem, where all products are unique: One-of-a-Kind Production (OKP). OKP problems represent those cases where the production process is focused on the customers’ needs and is tailored to their requirements. To address the planning of this type of problems, a CONWIP strategy was used to control production, avoiding overloads in the shop-floor. A particular detail of this work is that the productive configuration adopted is job shop, which is a non-traditional configuration for CONWIP strategy. Furthermore, to better represent the OKP nature of production orders, a great variability was introduced in the number of operations and time required by them. This feature hinders the CONWIP logic, since adding a new job to the shop-floor may not mean adding a workload similar to that of the job that left the shop-floor. To overcome this situation, 2 new dispatching rules that analyze the workload generated by each job before choosing the next job to be dispatched are proposed: SameOp-pure and SameOp-EDD. These rules consider the number of operations of each production order, and prioritize the jobs following a similarity criterion, dispatching as the next job the one most similar to the job that has just left the shop-floor and thus maintaining the workload inside the CONWIP at a fixed level. Computational experiments were carried out based on discrete-event simulation, and the benefits of these new dispatching rules could be verified, obtaining better results than other traditional rules (EDD, FIFO and critical ratio). It was even possible to verify that under more demanding scenarios, the advantages obtained by the SameOp-pure and SameOp-EDD rules were more significant.

The real-world problems often pose as multi-objective with competing objectives. Unlike single-objective optimization problems, multi-objective problems result in a large set of solutions called Pareto optimal solutions (Pareto set). All the solutions in this set are considered equally good with some trade-offs. Therefore, the decision-makers face the challenge of choosing a solution especially in the absence of subjective or judgmental information. On the other hand, analyzing all the solutions is not practical due to the time complexity. This means that a pruning method is needed to tackle this problem. Several methods have been proposed in the literature. These methods include clustering (e.g., K-means) and ranking (e.g., hierarchy process-based) of Pareto optimal solutions to reduce the number of solutions to a promising set with smaller cardinality. In clustering methods, a representative solution is extracted from each cluster to form the reduced set (e.g., the solution at the cluster center or one closest to the ideal solution of the cluster). However, the point closest to the ideal solution may not be a good representation for the entire cluster. Moreover, the reduced set may not contain the extreme solutions and, hence, does not capture the diversity of the entire Pareto set. Therefore, to alleviate the shortcomings of the existing approaches, we propose a novel graph-theoretical approach, which is based on the connectivity (e.g., degree) in the objective space, to obtain the representative solutions from each cluster. We test the applicability of the proposed method on the Pareto optimal solutions obtained from a multi-objective optimization model for a realistic case study. We show both qualitatively and quantitatively that the reduced set obtained from the proposed method better represents the entire Pareto set.

Currently, a common way of collaboration between product development and production planning is working with a Bill of Material (BOM) in terms of Concurrent Engineering, which is characterized by a structure as a traditional relational data model. Even though such a BOM can serve as a single point of truth of the product structure, it still leads to information discontinuities from upper- to downstream engineering teams, as they use BOM data for different purposes. Hence, a restructuring work of the product development oriented BOM is needed for tasks in production planning, so that the hierarchy is adapted. The parts structure needs to be rearranged to meet the requirements of the manufacturing and assembly processes. In the circumstances of volatile global market and varying customer needs, the BOM structured in relational data model provides insufficient digital consistency to reach an appropriate time-to-market and quick response on product changes and variations. In order to be able to cope with the increasing amounts of heterogeneous data, graph data model approaches are playing an essential role. Due to its non-centric information representation, it can be used as a neutral data model to dissolve boundaries in engineering. A graph model based BOM structure can fulfill both tasks in product development and production planning directly. Thus, it can reach an optimized consistency concerning information flow through various engineering activities as well. This paper proposes the graph model based BOM to minimize efforts and gaps in collaboration between product development and production planning. In addition, it also shows further potentials of usage in engineering.

The development and increasing application area of electric vehicles (EVs) in city logistics has contributed to the necessity of intermediate stops for charging EVs. Due to the limited service range of these vehicles, charging facilities must be located at the same echelon as customers’ and differ from depots or hubs in that they are visited while serving customers. Thereby, these facilities allow vehicles to exchange requests and get recharged. In this study, we present the Intra-route Location Routing for the Pickup and Delivery Problem with Transfers that arises in this novel scenario. There are sets of requests, vehicles, and potential transfer locations (intra-route facilities). Each request consists of a pair of a pickup node and a delivery node. Vehicles start their routes from their respective origins, serve customers (pickup and delivery), and return to the origin nodes. In the generic Pickup and Delivery Problem, a request must be served by a single vehicle that picks and delivers the request. In our study, however, multiple vehicles can service a request collaboratively by transferring it to each other at a transfer facility. In this problem context, we decide vehicle schedules, transfer decisions, and which nodes should serve as intra-route facilities. We propose a mathematical model for this novel NP-Hard problem. Experimental results indicate the computational difficulty of the problem in practice. For several small instances, solving the mathematical model by using a commercial solver does not even find the optimal solutions in six hours, and given the importance of this problem, it must be addressed in future works.

In recent years, the number of delivered parcels has rapidly increased along with the growth of the Internet and courier services have been booming. In addition, many people refrained from going out due to the spread of the COVID-19, and also people who have never used e-commerce have started using e-commerce. Therefore, it has been become to require to deliver parcels more efficiently. One of the evaluation measures to realize an efficient delivery is minimizing the burden of loading/unloading parcels. To do so, loaded positions of parcels should be decided by considering the unloading sequence of the parcels so that a delivery person moves few parcels at the time of unloading them. Hence, some systematic method for deciding loaded positions of parcels would be required. However, at present, the delivery person determines the loading positions of them with not a systematic method but his experience or intuition method. In this research, we would try to optimize loading positions of parcels by using genetic algorithm. Our proposed method of determining the loading position would use genetic algorithm with three-dimensional modified bottom-left method with multidirectional reference points. In addition, we would make comparison between the method of loading from only one direction and the method of loading from multiple directions. The number of planned movement of parcels would be treated as an evaluation measure as well as previous researches. We would make comparison between the precedent and the proposed method, and consideration of the results. We could conclude that our proposed method has dramatically improved the precedent methods. The results by the multiple directions were about 20% better than the ones by one direction.

Companies have different types of processes in their logistics chain. Thus, the added value delivered to each link generates a fundamental competitive tool, such as excellence and differentiation in the delivery service, coupling with technology and anticipating what customers need or expect, which means that companies must have an efficient inventory and resource management. On the other hand, companies need to procure goods and services for the development of their activities. These supplies are accumulated in the companies and must be managed for their correct handling and conservation. The problem that the SME has is that when customers buy in the store, in several cases, there is no product so sometimes the sale is lost, or they wait and return the day that the product is in the store. So in this work we study the case of an SME, companies that need inventory management to avoid losing customers. In this case, the inventory and sales process is modeled and simulated using Bizagi Modeler, where it was possible to obtain the breaks. Demand forecasting was studied with neural networks. Since the behavior of the demand is variable during the year, mainly December, February, Mother’s Day and others, which causes peaks in demand, to analyze this behavior we used the Periodic Review P Model, where we obtained the quantity to be requested to the supplier in different periods of the year, which increases sales, since 100% of the demand is covered. The results of the simulation with BIZAGE with the current and proposed situation show that in the proposed situation the loss of customers is between 0 and 0.5%. In relation to costs and profits, these were obtained with model P, although the costs increase by about 5% since a larger inventory is maintained than in the current situation, but customers will be 100% satisfied, which leads to an increase in profits of between 15 and 20%.

Chinese coastal megacities face an increasing challenge on urban floods in light of climate change and rapid urbanisation. Severe urban floods worsen damages of public infrastructures, such as roads and railway networks, because the land drainage system is currently insufficient to cope with intensive rainstorms. This condition is particularly severe in most Chinese megacities. As a result, that disrupted the transport system and logistics, which associated the interruption of economic and business opportunities. This paper took Guangzhou, the leading commercial and manufacturing hub in the Greater Bay Area region, as a case study. Unfortunately, the city is currently exposed to a high level of urban flood risk. By analysing the historical data, this research summarised the annual rainfall pattern in Guangzhou, and presented the potential impacts of the urban flood on logistics disruption based on the layout of road surface water flooding. Our findings in this research aim to provide recommendations on logistics disruption and transportation plan for road users in future flood events. Moreover, the results are insightful in tackling future urban floods to improve current policies on achieving resilient city planning on logistics.

Locating facilities has traditionally viewed from different disciplines. However, few studies are interdisciplinary. In this paper, we investigate the location of street markets in smart cities from two perspectives: urban planning and mathematical modelling. Smart city technologies can play a major role in the implementation of street markets: for example, smart street markets can show in real time the availability of products, market prices, etc. In this paper, we propose a mathematical model to locate “smart” street markets in the city of Bogotá (Colombia) with constraints that involve urban planning and public space. The results will show how this strategy can improve the current practice.

Producers are increasingly faced with greater challenges. The requirements in the manufacturing of high-tech products are constantly increasing, especially the demands on the production environment are becoming more relevant. The market demands change with high volatility. In addition, debates and subsequent changes in the laws on sustainable management and resource optimization are increasing, also based on growing public awareness. The concept of smart network production is a promising approach for industrial companies to address the resulting dynamics. Industry 4.0 measures allow production of smaller lots and higher manufacturing flexibility. Connected automation technologies and adaptable assisting technologies support these efforts. However, the flexibilization of manufacturing is only possible if the building offers suitable framework conditions. Factory buildings have a long-lasting life cycle even up to 30 years and more. Therefore, the aspects of optimization of energy consumption and high flexibility of the production areas need to be considered in planning projects, too. Additionally, the support of digital production and smart building maintenance need to be planned in an early stage. To reduce the impact on the surrounding, the integration of factories into the environment as well as the definition of symbiosis with urban infrastructures need to be considered in ultra-efficient production and innovation areas. In the context of the article, it shall be discussed how current trends in digitalization of the production affect the planning and realization of building processes, as well as the question, which requirements need to be considered to realize production infrastructure serving future needs. An important factor to realize energy efficiency and sustainability does not only cope with the building itself. The integration of aspects concerning factory surroundings are also important in supporting acceptance for new constructions and production units. Compliant production and offerings by local enterprises increase the positive social impact on the surroundings. The analysis of a pioneer company in the field of factory building construction enables the evaluation of economic business development combined with sustainable action. The implementation of currently existing technology support fulfils the objective, generating a positive impact along the lifecycle of the building.

Green supply chain applications are gaining wide interest as environmental and economic concerns have been increasing. Transportation authorities impose strict regulations to reduce the negative effect of transportation activities on the environment such as carbon emissions and noise. In this study, we introduce a novel optimization model for a green multiple allocation hub covering problem that seeks to find the best locations for hubs and allocation of origin/destination flows in a logistic network. Different from the existing hub covering flow problems, carbon emission costs is added to the total cost function aiming to minimize the total amount of emissions in the network. Further, we investigate the effect of several factors (e.g., coverage radii of hubs) on optimal solutions. To illustrate the efficiency of our approach over the conventional hub location models, we first generate a new dataset by combining a well-known dataset used in traditional hub covering problems with a carbon emission dataset provided by the International Air Transport Association (IATA). Afterward, we present a comparison of our model to previous models that do not consider carbon emission.

Food production is an essential operation where production and resource efficiency are low compared to other sectors. With the unstoppable growth of the world population, agricultural production is under pressure to meet the increasing food demand. Controlled Environment Agriculture (CEA) is a successful solution to create sustainable and resilient development through sustainable cities. CEA, where the farming activities are isolated from the meteorological conditions, is one of the most powerful solutions to adapt and mitigate climate change in urban areas. Vertical farming (VF) is also an indoor plant manufacturing process. In VF, plants are grown in layers and can thus reach high. The system can entirely be designed without any dependence on sunlight or other outdoor resources. However, there are a significant number of drawbacks about VF in the literature, such as limited products and labor costs, etc. This study focused on generating the VF area’s main challenges and wanted to create a roadmap to overcome these challenges. Existing VF challenges are gathered from experts and related literature. Possible solutions to overcome these limitations are derived from the literature as well. The process is approached as a multi-criteria decision-making (MCDM) procedure. The House of Quality (HoQ) of Quality Function Deployment (QFD) is suggested to investigate the relationships between solutions and challenges. The HoQ method also allows for prioritizing the potential solutions to generate a roadmap for practitioners. Plus, the methodology extends the QFD model with the 2-tuple linguistic model to overcome the vagueness by supplying linguistic sets to decision-makers (DMs) to assess via semantics closer to the human cognitive process. That helps to improve the accuracy of the linguistic computations and interpretability of the results. Also, it creates a flexible environment for DMs. A case study is applied for Turkey, and sensitivity analyses are presented to test the suggested methodology’s robustness.

The current study would like to assess the perception of the staff involved in new product development, while the research investigation was conducted by a questionnaire-based survey configuration with the corresponding data analysis. Practically, the study provides the assessment about the perceived difficulties of the stages of the new product development process, namely, idea generation, idea screening, concept development and piloting, development of marketing strategy, business analysis, product development, testing the market, and commercialization. Along with the general demographic information of the engineers, including age, gender, level of education, rank, and years of service, the research questionnaire was composed to collect the required data for the exploration of the relationship between them can be revealed. With the self-defined research instrument, 132 valid responses were collected for the modelling process. After the completion of the analysis, testing the market was found to be the most difficult, and the particular difficulties for concept development and piloting and business analysis can be observed from the junior staff. With respect to the situation, the recommendations were provided in the latter part of the paper.

In the 24th ICPR, we unraveled the unique Chinese innovation pathways and clearly explained its distinctiveness from the Japanese and South Korean. In this paper, we examine SMEs innovation pathways using empirical data gathered in Vietnam. Under current America’s ‘decoupling’ strategy, Vietnam will benefit greatly from manufacturing firms moving out of China. Vietnam is now a major destination of foreign direct investment in research and development and an attractive knowledge-based location for leading MNCs. Many MNCs (such as GE Healthcare, Philips, Siemen, Hitachi, Mindray) have set up their R&D units in Hanoi and/or Ho-Chi-Minh. Hence, a better understanding of Vietnamese innovation in the context of Vietnam’s emerging economy, evolving institution, and growing firm capabilities is beneficial to practitioners, policy makers and academia. We explore the unique innovation pathways from the perspective of Vietnamese medical equipment manufacturers (MedTech) and provide insight from managers on how international firms could galvanized the uniqueness to their product development advantage. The results provided interesting insights into how MedTech (a high growth sector in Vietnam) SMEs cultivated their data analytics, collective and target innovation, digitalization skills and know how in various innovation phases. Vietnam is not a paragon of innovation at all, but it certainly has learned how to invent sustainably with less, and that is something other emerging and Asian economies can learn from. We believe the findings from our empirical case studies can contribute towards a better understanding of how Vietnam has been able to evolve from a position of technology borrower to technology innovator. Moreover, findings from this research help to shed light on existing Asian innovation development debates.

Digital collaboration in all sectors of industry is on the rise, also driven by the COVID-19 pandemic. People centricity, location independence and resilient delivery are important trends in this regard. Crowd engineering is a possible solution to address the challenges posed by these trends. This paper provides an insight about a technological approach for a platform for community-based product engineering. The platform consists of different cornerstones to ensure flexibility towards users, open interfaces and transparency to support re-use of existing project results. As different users have different requirements, companies have over time grown an IT infrastructure serving their needs best. Prosumers, i.e. consumers which are involved in the production of the respective products, have a limited set of digital tools allowing them to contribute to product creation. A suitable platform should be able to integrate different approaches and data from various digital tools. Rather than being monolithic, an appropriate platform needs to be a hub for core data and functionalities to support secure and reliable collaboration between all involved actors, corporate or private. These active users will be organized in a virtual community to create an attractive platform for the exchange of ideas, concepts and realizations for technical solutions by developers. The paper will introduce a possible platform structure where personal data, project related data, development data, technical product descriptions and community interactions are combined. A common access point allows an aggregation of touchpoints without centralization. Apart from the examination of technical and structural aspects of such a platform, approaches for a possible operator model are discussed. In addition to providing technically required functions, a lively community is an important basis for successful crowd engineering. Therefore, this paper will examine both community building and the implementation of co-creation projects, as they are necessary to operate the crowd engineering platform economically. Finally, different potential further advancement of the Crowd Engineering platform shall be discussed. This focuses on technological solutions that support users or serve certain tasks.

Opening up product development in companies to external interdisciplinary participants offers the opportunity to develop products faster and more cost-effectively. An early integration of potential customers in the product development process enables a stronger focus on their needs and thus facilitates customised product development. A lack of suitable tools and rigid processes within companies, however, make this integration unpractical in many cases. In this context, Crowd Engineering allows the involvement of an engineering community, with community members being integrated in such a way that they can participate actively and in a distributed manner in product development using a central internet platform. Meanwhile, artificial intelligence methods and approaches are becoming more sophisticated and acquire more importance in various aspects of private life and business context by providing new solutions to topics which formerly were not solvable or only with high efforts. This paper analyses how artificial intelligence can support product development, especially in the context of Crowd Engineering. It thereby provides an overview of how artificial intelligence already supports development processes today and which potentials arise from this. Furthermore, an analysis of the collaboration between humans and artificial intelligence is carried out in the context of Crowd Engineering. The main research questions are how artificial intelligence can support in collaboration and how the role of humans is being shaped and transformed. Another key question is how collaboration can be achieved in the context of Crowd Engineering—hence this paper proposes collaboration modes that vary according to three dimensions of collaboration: degree of AI-assistance, degree of task complexity and the degree of AI-autonomy.

The innovation process of manufacturing companies is characterized by uncertainty caused by various external influences. Today, companies are not able to handle the amount of data available for the identification of changes within their products’ environment although, this data can give valuable indications for the urgency or possibility to innovate their products. Approaches from environmental scanning, text mining and semantic networks have the potential to address individual aspects of that problem separately but there is no method for monitoring the environment holistically. The presented framework closes this gap by combining methods from the three areas that complement each other for the identification of potential fields of innovation for existing products. The aim of this paper is to enable companies to process text data from external data sources automatically and therefore generate insights for the detection of potential fields of innovation within the environment of their products. For this purpose, the product is described as a semantic network. Further, relevant external influences, i.e. customers and competing companies or respective external data sources for their description are identified. A text mining approach extracts the topics covered by these data sources and expands the product specific semantic network by linking them based on their co-occurrence with each other and the product description. Finally, the potential fields of innovation are identified and evaluated in terms of external relevance and the company’s competence.

To come up with new ideas and innovations, product developers are faced with the challenge of finding a suitable way through a jungle of tools and methods, e.g., creativity tools, CAD software and project management applications. The goal of this paper is to develop a user interface concept for a supportive tool that integrates a variety of tools on an interactive surface to enable horizontal collaboration and interaction during creativity sessions in product development. Therefore, the user interface should not just support collaboration but also incorporate guidance to support product developers to find a suitable path through this jungle of tools and methods. To elicit requirements for such a supportive tool, an explorative multi-stage study design was chosen. In the first part expert workshops were conducted to create a design space. This space was further explored in a second phase, to derive relevant requirements. In the last phase, first visions of the supportive tool and possible design ideas were developed. The results of this study form a foundation to build a supportive tool for product developers that could pave the way to a new form of digital collaboration, provides a more holistic understanding of the development process and its activities and furthermore enhances the acceptance of creativity methods.

Because of a high product and technology complexity, companies involve external partners in their research and development (R&D) processes. Interorganizational projects result, which represent temporary organizations. In these projects heterogenous organizations work closely together. Since project work is always teamwork, these projects face due to their characteristic’s major challenges on an organizational, relational, and content-related collaboration level. Thus, this paper raises the following research question: “How can a project team be supported on an organizational, relational, and content-related level in an interorganizational new product development setting?” To answer this research question, an explorative expert study was set up with two digital workshops using the interactive presentation tool Mentimeter. The results show that a cooperative innovation culture could support project teams on an organizational and relational level in the future in minimizing predominant problems. Moreover, it supports project teams for example in a functional communication. Furthermore, 18 values of a cooperative innovation culture result which are for example openness and transparency, risk and failure tolerance or respect. On a content-related level the results show that an adaptable tool which promotes creativity and collaboration method as well as content-related input support could be beneficial for problem-solving in an interorganizational new product development setting in the future. Because the tool can guide product developers through the process with suitable creativity and collaboration methods, can give content-related input and can enable interactive interchange on a table-top. Future research could mainly focus on the connection of the cooperative innovation culture and the tool since these potentially influence each other.

Whereas scholarly studies often elaborate on the design of the method for empirical data collection and analysis, surprisingly, the actual use of design principles taken from the domain of design and engineering of products has not come into the picture, yet; the stance of this paper is that the use of design principles taken from this domain may be beneficial for making choices and detailing research design. The application is demonstrated through showcasing the research design process of a doctoral study. The topic explored in this doctoral study concerns the theoretical concepts of exploration and exploitation, especially focused on their implication in managerial practices of managing innovation. In addition, both theoretical and empirical validity of view exploration and exploitation as a dichotomy was questioned; this was done through challenging the domain assumptions sustaining these notions. In order to tackle this rather complex research problem, the study adopted Pugh’s ‘controlled convergence method’ to have a more systematic approach in setting the final research methods.

Markets are increasingly demanding greater product variety and, at the same time, ever shorter delivery times. In the case of new product development, it is often no longer possible for production to ensure a sufficiently high delivery capability with the help of lean methods and value stream methods. Especially if acceptable inventories and costs are assumed. During product development, products and product systems have not yet been sufficiently coordinated with each other from a value stream perspective and designed in a targeted manner. For this reason, it is crucial to align the product and its future production system with the goal of a functioning value stream as early as the concept phase of product development. Up to now, the value stream methodology has been used to analyze and optimize real production processes that are already running. In the future, the delivery times and delivery reliability of new products are to be secured by simulating the value stream based on the value stream methodology. In the context of Advanced Systems Engineering (ASE), the value stream resulting from the product is already created in an early phase of product development in order to be able to check the functionality of the production system and the deliverability of the product. Value stream simulation makes a significant contribution to improving cooperation between product development, production and logistics, as we can vividly present in this article using the ASE demonstration product. The value stream simulation revealed that the demonstration product had design weaknesses in terms of the required delivery capability. Assuming that no changes to the pro-duction system are possible, design changes were made to the product. The housing, which runs through a bottleneck line with over 100 variants, was standardized and the number of variants reduced to two. This led to later variant creation in the value stream. The variants now only appear in assembly, which can cope with many variants due to its high variant flexibility. The simulation of the revised product and the new value stream resulted in a significant improvement in delivery capability and the required short delivery times to the customer can be met.

As an alternative to nonconforming product rate used traditionally as an index of quality, Taguchi has proposed quality loss. From the viewpoint of this new index of quality, some variable sampling inspection plans for assuring product quality indexed by quality loss have been proposed so far. The quality loss is defined as the expected loss by derivation from a target value of product quality characteristic. In concrete, the quality loss is composed of both of the square of the difference between the mean in the actual product quality and the target product quality, and the magnitude of variance in the actual product quality. And then, these two components have the same magnitude of effects in the evaluation of quality loss. On the other hand, it is known that for improving product quality at a production site, adjusting the mean in product quality is often easier than lessening the variance in product quality. Hence, we can see that adjusting the mean is more efficient and effective than lessening the variance as a way of reducing the quality loss. From the mentioned above, this research will consider a single sampling plan indexed by quality loss with consideration of emphasizing the loss derived from the mean in the product quality. And, the operating characteristics of the proposed sampling plan are confirmed through some numerical simulation.

Variations of component quality have a significant impact on product quality and its performance. The variations in the manufacturing process are due to variations of dimensions and geometrical features caused by machining errors. The selective assembly method is a standard method for improving product quality. However, it takes time and costs much because all products should be measured, classified in rank, assembled into a product. In this study, we proposed a method to stabilize product quality by managing the parts that are produced machined in a later process. The effectiveness of the proposed method is verified in simulations using a virtual product consisting of two parts. Following two cases were simulated to verify the effectiveness, Case A where the process mean is randomly set assuming a situation where two parts are made individually, and Case B where the process means of the part machined later is controlled according to process means of another part machined before. The cases A and B are compared by calculating values of process capability index Cpk calculated by the process mean and standard deviation when the two parts are assembled. In order to visualize the fluctuations of the process mean and process capability index, control charts of Cpk for each lot and investigated the Cpk statistics. The results showed that the proposed method could significantly decrease a variation of process mean and implied that specifying limits of process capability index in design drawing can realize quality design in mass production.

The rational inattention model has recently attracted much attention as a promising candidate to model bounded rationality in the research field of decision-making and game theory. The model presumes that the cost of information processing is proportional to the mutual information obtained from signals. It has been reported that the introduction of this information cost can explain various phenomena observed in the market, and applied in a wide range of fields such as finance, bargaining, auction, and policy analysis. However, the rational inattention model does not have a sufficient cognitive foundation, despite the amount of attention it has received. In this study, we will use both a behavioral and neural approach to investigate whether the amount of mutual information corresponds to cognitive costs.

3D printers are a fascinating tool for engineering teaching and education, related to a large variety of subjects. More and more universities are integrating not just the topic of additive manufacturing, but 3D printers’ infrastructures to create great learning experiences. The main scope of the paper is to analyze the state of teaching and learning 3D printing in the universities, presenting the advantages, particularities, and the involved actors. There will be supported the ideas that through 3D printing, students can translate their ideas directly into reality, and spatial imagination. Usually, initial teaching and learnings activities begins with simple physical objects and later deals with abstract, virtual 3D models and complex assemblies. The “magic” of teaching and learning 3D printing is that it allows quick reversal, from the 3D CAD drawing to the physical object; the direct link of the two processes is stimulating creativity and enhance imagination. Finally, there will be discussed the case of teaching and learning 3D printing at Politehnica University of Timisoara (Romania) with the support offer by the “3D Printing Support Service for Innovative Citizens” INNO3D project (2019-1-IE203-000693INNO3D).

The paper aims describing a collaborative international initiative of four European universities to develop a training program for supporting university teaching staff in creating and using multimedia technologies effectively (extend their digital competencies). Thus, teaching and researching staff from higher education institutions will follow a designed framework to develop their digital skills to create more realistic and attractive content that should improve engineering education (with a debate based on examples and best practices of different specializations). We relate the presented approach to MUST project (Multimedia Competencies for University Staff to Empower University—Community Collaborations, 2020–1-RO01-KA203-080399). The training curriculum and the created educational resources make up a dedicated service offered by universities through DigiCoaches who will provide training to other internal/external users/trainees in creating-using multimedia technologies effectively. The paper will enhance the quality and relevance of knowledge and skills of university teaching and researching staff in multimedia technologies by presenting an overview of the training needs assessment study. Five universities from Romania, Spain, Lithuania, Slovenia, and Republic of North Macedonia with two consulting companies from Germany and Portugal (experts in creating digital resources) will support the new multimedia curriculum and training programme development.

Purchasing additional MRI machines may not always be a feasible solution to increase diagnostic services capacity. Healthcare systems make use of appointment scheduling to optimize the use of their limited resources as an alternative to investing in new machines. A properly planned appointment scheduling system will allow for the efficient use of the MRI machine and accommodate patients effectively. The challenge in appointment scheduling comes in the various uncertainties that could disrupt the planned schedules. Current research and literature on appointment scheduling for MRI scans have not tackled the appointment scheduling problem while considering the occurrence of unpunctual patients, no show patients, and unscheduled arrivals, at the same time. This study develops a scheduling model that considers the aforementioned uncertainties in addition to different patient types. A two stage solution is utilized in order to create a weekly appointment schedule. The first stage determines the total number of patients to be accommodated for the week. This is then used as an input for the second model to determine the time slot assignment of patients. The latter minimizes the total penalty contributed by the priority levels and the cost of rearranging from the original order of requests. A neighborhood search heuristic is used to develop a prioritization rule when such uncertainties are realized. From the results and analysis, it was found that the model returns a schedule that generates a lower value of the objective function compared to ad hoc scheduling practices such as that of scheduling according order of request only or according to priority level only.

Around the globe, pharmaceutical companies are increasingly becoming connected, internally as well as externally. Data and information is significantly important to deliver better quality healthcare. A large number of pharmaceutical business running legacy systems will not survive without investing in technology, data systems and innovative models. Hence, innovative business models warrants better product mix with complex generics and specialty products. We identify how supply chain management systems help in coordinating the planning and operations of emerging business models in pharmaceutical sector though network optimization. Since pharmaceutical products are for human consumption, complete control over distribution chain is mandatory. Hence, we focus on modern supply chain having traceable, secure, globalized and compliant logistics for the pharmaceutical sector. We simulate the pharmaceutical supply chain using digital twins. We utilize discrete event simulation methodology using anylogistix simulation and optimization toolkit. The case study in this work is the Jan Aarogya (peoples’ health) scheme in India, which is built on the modern supply chain paradigm of hub and spoke architecture. We find that digital twins can significantly improve the decision-making during large-scale healthcare disruptions. We simulate pharmaceutical supply chains during disruptions by focusing on system resilience. A system resilience is composed of component resilience, buffer inventory and time. We derive the impact of disruptions and build a simulation model for the located pharmaceutical hubs. Based on our simulation model for pharmaceutical supply chains, we show the impact of network utilization and proactive planning. Suitable recommendations for policy makers is also highlighted based on the analysis results.

This paper presents an integrated approach for the dual resource (medical staff and emergency beds) scheduling at each Point of Care (PoC) under time-varying Patient arrival demand for trauma centers. Practical constraints, such as the number of beds available, the number of patients that can be treated simultaneously at each Point of Care (PoC), and the signing on/off medical staff, are considered. The objective is to reduce the Patient’s waiting time before entering into Operation Theatre/discharged since the Patient’s arrival to the hospital. The mixed-integer linear programming (MILP) technique is used to solve the resulting multi-objective to deliver medical staff scheduling (i.e., signing on and off times of medical staff) and bed requirements simultaneously. The reduction of waiting times with this proposed algorithm is analyzed. A case study is conducted based on real-life data from a trauma center in India. The proposed approach is compared with the practical approach being followed in the trauma center. This comparison shows the proposed approach’s effectiveness, suggesting scheduling medical staff based on historic patient arrival patterns will help reduce Patient’s waiting times before entering Operation Theatre since the patient arrival. There is an average improvement in total process times for patients by 30%, with enhancements considered at a critical PoC.

Previous studies usually investigated improve the overcrowding in the emergency department (ED) as the research goal. In the case of limited medical resources, the congestion in the ED was due to the capacity of the hospital, and the imbalance of resources in the hospital is one of the bottlenecks. For this problem, we simulated the use of bed resources in different departments through discrete events, and dispatched the allocation of bed resources in real time. Finally, we compared three bed resource scheduling plans. This study shows that the best plan can improve the proportion of admitted patients held >48 h from the original 5.08 to 1.83%; The proportion of admitted patients held >24 h, stays dropped from 19.49 to 16.21%, and the variance of daily ED occupancy was reduced by 7%.

The Philippines is one of the populous nations in the world that suffers from the scarcity of healthcare resources. The research aims to increase the percentage of population accommodated by healthcare resources in barangay health centers annually using Integer Linear Programming (ILP). It will be used to address the problem of the insufficiency of healthcare resources in the Philippines which can also be used on different local barangay health centers. The study will be comprised of two models—the Human Resource Model and Basic Medical Equipment Model. The researchers also conducted three different cases for each model to test the feasibility and optimality of the model. For the first case, the base model, it is found that there is an infeasible solution with the data used from 2018 for the Human Resource Model. For the second case, decreasing the budget of 17% given to the Department of Health for 2019. This case decreased the budget allocated for the Human Resource of healthcare resources which makes the model more infeasible. For the third case, the surplus in the budget constraint of the Basic Medical Equipment Model will be used as an incremental for the budget constraint in the Human Resource Model to point out the lack of budget as a source of infeasibility for the first two models of Human Resource.