Intelligent Engineering and Management for Industry 4.0

This research deals with a three-stage flowshop, processing two types of products. In the first stage, two heterogeneous machines assemble electronic components, and in the second stage, two dedicated parallel machines assemble the assembled electronic components with other components to produce respective types of products, where both the stages constitute job processor machines. The third stage is a common machine of a batch processor to conduct a functional test. The objective is to minimize total actual flowtime, where the actual flowtime of a part is defined as the time interval between its starting time of processing and a common due date. The problem is formulated as a mathematical model, and a heuristic procedure consists of two algorithms to solve this problem is proposed. The first algorithm is used to determine production batch sizes and the batch production sequence, whereas the second algorithm is used to select the machine in the first stage. The proposed models and algorithms are built by integrating all processes that occur at each stage. An illustrative example is intended to show that the proposed algorithms could perform well.

A three-stage hybrid flow shop consists of machining, assembly, and differentiation stages processing different product types, each of which is considered as a job. The machining stage processed the unique and common components of a job on unrelated parallel machines. The unique component is specific for a certain product processed individually, while the common components are the same for all products processed in batches with a setup between consecutive batches. The assembly stage starts when the components of a job are available. The differentiation stage consists of dedicated parallel machines processing types of products. The problem is to determine an integrated schedule of the three-stage hybrid flow shop system processing different product types considering multiple due dates with the objective of minimizing total actual flow time. The total actual flow time allows parts to arrive in the shop at the starting time of processing and deliver the completed products at their respective due dates. This problem is formulated in a nonlinear programming model and solved by a proposed algorithm adopting an SPT-based heuristic for the initial solution and a variable neighborhood descent method with insert and swap move operators for optimizing the solution. An illustrative example is used to test the model considering multiple due date conditions. The result shows the ability of the algorithm to solve the problem.

Intractable problems in a job shop are studied. The shop is dual resource constrained and all workers are skilled and thus eligible to operate every machine. A worker may be able to move away from a machine that is in operation to support the process on another machine. This condition indicates that the operation of a job on a machine is composed of manual work units and automated work units. A manual work unit on a machine can be processed only when a worker is allocated to the unit. As the assigned working hours of workers involve break times, the manual work unit will be interrupted during a break. On the other hand, automated work units can be operated even during a break and outside working hours. In other words, automated work units can be scheduled without considering the presence of workers. Because of the availability of alternative machines, and other complicated factors, interactive scheduling is a practical and useful approach to generating an acceptable schedule. An interactive scheduling system tailored for the shop in focus is illustrated. Techniques that reduce the computational burden to improve the responsiveness and several remaining issues are discussed.

Almost all key techniques of Cloud Manufacturing (CMfg) refer to services and combination of services, which make it urgent to deeply explore intrinsic characteristics together with evolution rules of relationships between services. As a critical topic of CMfg, service combination and optimal selection (SCOS) will also benefit from the exploration. Hence, the feature of similarity between services is studied with the method of network analysis and applied in services importance evaluation and clustering. A similarity evaluation method for CMfg services is proposed firstly based on service invocation history. Then, a service similarity network model is established and visualized by means of a similarity adjacency matrix. Furthermore, importance of each service is evaluated by three characteristics of the service similarity network model, i.e., degree, eigenvector centrality, and clustering coefficient. A case study validates the feasibility and effectiveness of the proposed similarity network model together with related similarity evaluation method.

This study focuses on system reliability evaluation methods for the linear connected-(1, 2)-or-(2, 1)-out-of-(m, n):F lattice systems. To compute system reliability, previous studies have proposed the recursive equation and finite Markov chain imbedding approaches. However, no study directly compared the efficiency of both approaches. Hence, for the first time in the literature, this paper comprehensively compares the efficiency of both approaches. Accordingly, infer that their efficiency depends on the situation. This comparison will provide a guideline to help decide the approach that should be adopted. Furthermore, upper and lower bounds are also employed, provided it is unnecessary to obtain the exact system reliability. Several useful and simple bounds were reported; however, tighter bounds that require the computational burden have not been sufficiently discussed. Therefore, we derive the upper and lower bounds for the system reliability. The results of the numerical experiment indicate that we obtained the tighter bounds at the expense of computational effort, compared with existing bounds. These tighter bounds can sufficiently evaluate the reliability of large systems. The theoretical results in this paper would provide useful insights into industrial manufacturing, contribute to the stable operation of practical systems, and promote Industry 4.0.

On traditional assembly lines, each worker is usually assigned to a particular fixed operation. However, when there is an imbalance between the speeds of workers, a particular worker delays the overall work on the line, and the production rate of that line will also decrease. To avoid this problem, the “self-balancing production line” was introduced. On this type of line, each worker is assigned work dynamically, and balanced production can be maintained in this way while satisfying certain specific conditions. In previous research, a speed that depended on both the worker and the station was considered for a serial line. For these conditions, a worker coordination policy was proposed that changed the worker sequence based on the average speed. Using this policy, the production rate may decrease, and blocking by a slower worker, which may hinder the work of a faster predecessor, is not considered. As a result, balancing of the line cannot be achieved. Exact solution methods are therefore needed, and in this chapter, an exact solution method for balancing the line is proposed. We also present numerical calculations for two and three workers in order to enable a comparison with the worker sequencing policy put forward in the previous chapter.

A dual resource-constrained (DRC) scheduling problem of identical parallel semi-automatic machines considers a fewer number of operators who could control setup and unloading tasks. This problem allows the operators to move between machines on the machining time. The objective functions are the makespan and workload smoothness index (WSI) measuring the workload balance between operators related to the social topics in Global Reporting Initiative Reporting Standards (GRI Standards). This problem is complicated because the schedule contains four information, i.e., machine and operator assignments with job and task sequence.

This study proposes a novel Bi-encoded Non-Dominated Sorting Genetic Algorithm II (BNSGA-II) as a developed version from the existing Non-Dominated Sorting Genetic Algorithm II (NSGA-II) for solving the DRC scheduling problem. The results show that BNSGA-II equipped with an additional task chromosome could contribute more solutions that NSGA-II could not achieve in the Pareto front for the small and medium-sized problem. It also could generate some new good solutions in the large-sized problem.

Our paper discusses optimal trading strategy of stock using limit orders considering nonexecution risk for institutional investors. The limit order could satisfy their needs due to the smaller market impact than market order. However, the limit order has risk of not being filled which is called nonexecution risk. According to some empirical analyses, executing a larger amount of limit order is more difficult than a small amount, and this relationship should be considered in the execution strategy. We estimate the execution probability distribution empirically. The reorder strategy proposed in our paper allows investors to replace nonexecuted limit orders as new limit orders, and nonexecuted amounts at maturity will be executed through market orders. The strategy is determined considering the trade-off among nonexecution risk, market impact, and timing risk. We find the optimal strategy can reduce the execution cost with the nonexecution risk.

Globally, financial inclusion has been high on policymakers’ agenda. The large population in some of the developing countries makes the task increasingly difficult. The recent developments in financial technologies are likely to provide the much needed low cost, efficient, and effective financial services that have alluded the large unbanked population for many years. This chapter evaluates the relative performance of Indian states for a unique financial inclusion initiative by the government of India which utilizes no frill bank accounts, unique identity authentication and mobile technology. The results so far appear encouraging and efforts need to continue to reach the goal. The chapter concludes with the implications of the findings for government, policymakers, bankers, regulators, and mobile companies.

Much attention and resources have been put to the integration of emerging technologies such as robotics, Internet of Things (IoT), big data analytics, Artificial Intelligence (AI), and cloud computing, for developing smart factories. Yet, the key to success in the Industry 4.0 era depends not only on the advancement and adoption of intelligent engineering for production but also on a concerted effort of intelligent systems management applied across multiple functions within a company and multiple partners on the supply chain. To show the potential benefits of applying an intelligent systems management approach to inventory management, an adaptive intelligent redeployment strategy is developed to integrate replenishment and redeployment of excess stock strategies with the application of redeployment model in a closed-loop service logistics network. A case study is presented to illustrate how an international high-tech company can apply this model and provide better customer services at lower costs. This adaptive strategy compels managers to rethink the conventional way of managing inventory of items with non-stationary demand and to pursue digitalization of inventory management jointly with supply chain partners for operations excellence in the Industry 4.0 era.

Recently, in Japan, the importance of elderly daytime care facilities which provide nursing care services has been increasing. Rehabilitation services such as massages by licensed staff and exercises using equipment and machines are offered in these facilities. Multiple staff members execute the rehabilitation services according to a staff schedule. However, the staff in the facilities often face problems of heavy workload. The workload problems of the staff were not grasped earlier since analysis of the workload based on the industrial engineering (IE) method was not conducted in the facilities. Therefore, there is a possibility that the workloads of the staff are not well-balanced in the schedule. In addition, it is considered that the staff have two types of workloads: physical and mental, since they offer their services to human beings instead of working with products and materials in factories. Thus, it is necessary to consider the mental workloads on staff in planning staff schedules in order to ensure that they are well-balanced. This study addresses a staff scheduling problem for elderly daytime care facilities considering mental workloads of the staff and plans a balanced schedule to incorporate the mental workloads. A balanced schedule for mental workloads among staff members is planned based on the workload survey that included an interview and questionnaire, conducted on actual staff members. The results are discussed by comparing the findings with the current schedule and the balanced one for physical workloads.

Social performance is a measurement of organizational outcome in social, environmental, economic, and government domain by considering multiple stakeholders. Increasing the social performance of Small and Medium Industry (SMI) can be done through innovation. Innovation concept that is suitable to be implemented in SMI is open innovation. Open innovation is characterized by the existence of knowledge flow among SMIs and between SMI and stakeholders. Knowledge flow in an organization is managed through knowledge management. The relationship between knowledge management and open innovation has been discussed in previous literature. However, there is no previous research that identifies the role of knowledge management and open innovation which supports the enhancement of SMI’s social performance. In this study, a conceptual model which represents the relationship between knowledge management and social performance through open innovation in SMI context is developed and tested. The knowledge management concept is represented by knowledge identification, knowledge acquisition, knowledge retention, knowledge sharing, and knowledge application. The open innovation concept consists of two dimensions which are inbound and outbound. The social performance concept is represented by the focus on community and employee. The respondents of this study are 27 small and medium industries which produce batik in Indonesia. The model is tested using the partial least square. The result shows a significant relationship between knowledge acquisition and inbound open innovation, knowledge sharing and outbound open innovation, and inbound open innovation and social performance.

Today’s engineering contractors face increasing uncertainty with EPC (Engineering, Procurement, Construction) projects because of increased project complexity and scale. Due to these circumstances, the number of joint venture contracts has increased among EPC contractors to reduce risks and increase profits. In this chapter, a method to design a competitive joint venture formation in consideration of cost and risk reduction by complementary effects among joint venture partners is developed. In the design method, the 2D-WBS (Two-Dimensional Work Breakdown Structure), which is associated with the work packages of the project to the potential partners carrying out the work packages, is used to structure project data of each partner organizing the joint venture formation. The design method uses a mathematical model to identify the candidates of the joint venture formation that minimizes the variance of the estimated project costs under the constraints of the expected project costs. The method then selects a joint venture formation that maximizes the expected profits by using a simulation model of competitive bidding. The effectiveness of the joint venture is demonstrated via simulation experiments using a simulation model of competitive bidding. As a result, it has been found that the joint venture contract can improve profits of contractors as well as reduce investment costs of clients.

This chapter aims to provide references of the collaboration impact to blanket order replenishment system by performing dynamic system simulation. System Dynamics Simulation has been widely used in understanding systemic implications throughout the supply chain. Two different types of relationships in the industry, sister company and brokerage, are tested with several scenarios of system dynamics stock and flow model that are created in Vensim PLE-Plus based on beer game modification. For each type of industry, one scenario is developed to mimic the current system, and some other scenarios describe alternatives of improvement. For both types of industry, the implementation models involve a three-echelon supply chain. The simulation reveals that collaboration in blanket order is able to reduce the bullwhip effect and total supply chain cost. The case studies demonstrate the best scenario of collaboration in a blanket order system is a retailer–supplier partnership with minimizing delay. The best scenario has reduced cost and bullwhip effects. As a continuation of this strategy, a managerial implication had been developed by empowering the strength of blockchain information sharing.

Biomass co-firing in coal power plants is an immediate and practical approach to reduce coal usage and pollutant emissions because only minor modifications are required. With direct co-firing, biomass can be used directly as secondary fuel in power plants to partially displace coal. Although it requires minimal investments, it can lead to equipment corrosion from unconventional fuel properties of the biomass–coal blend. With indirect co-firing, the risk of damage is minimized by separately processing biomass. The solid biochar by-product can be used as soil fertilizer to achieve further reductions in GHG emissions through carbon sequestration. However, as this calls for a separate biomass energy conversion plant, its investment cost is higher. Moreover, this system faces uncertainties from the inherent variability in biomass quality. This must be accounted for because mixing fuels results in the blending of their properties. In this work, a robust optimization model is proposed to design cost and environmentally effective biomass co-firing networks that decides on appropriate co-firing configurations and fuel blends. A case study is solved to demonstrate validity. Results of Monte Carlo simulation show that the robust optimal network configuration is relatively immune to uncertainty realizations as compared with the optimum identified with deterministic models.

Computer-aided design (CAD) systems have grown into a very important part of the design process, especially for the design of electro-mechanical products. On a closer look, it can be seen that CAD technologies are used extensively during the detail design phase of product design, while their use during the conceptual design phase still remains minimal. In this chapter, we propose a conceptual model for a CAD system to be used in the conceptual-embodiment design phase of product design. It is designed by taking into account the 4C concept of future CAD systems: conceptual, collaborative, creative, and cognitive. Based on the cognitive co-evolution design theory, the proposed system is designed such that it would be able to handle exchange of ideas between designers during the conceptual design and embodiment design phase while also maintaining the relationship between product functions and its physical embodiment.