Complex Systems Design & Management

Model-Based System Engineering (MBSE) and Product Line Engineering (PLE) are well-known approaches in industry for the management and design of the architecture of complex systems. The railway signalling business has some specific characteristics that need to be considered in system engineering: railway signalling systems have a long life time and new systems have to integrate interfaces to many types of legacy railway safety products. This situation has led to different technical system approaches: railway infrastructure companies as customers prefer either turn-key projects fulfilled by one supplier or tend to define individual subsystems that can be integrated to a complete system. This article shows how Thales masters both approaches by using the method ARCADIA and the open source modelling tool Capella in the specific case of pre-existing subsystems and how the resulting variability is handled. An outlook will be given to extensions that allow an early safety analysis of models and will provide support for automatic test design.

The Trans-Alaska Pipeline System (TAPS) was constructed between 1974 and 1977 in response to the 1973 oil crisis. It conveys oil from Prudhoe Bay in northern Alaska to the port of Valdez in the southeast, a distance of over 800 miles (1,300 km). Building the pipeline system meant dealing with a multiplicity of complex design and management decisions that involved engineering, environmental, political, legal, security, financial, and other issues. A decision was made to run most of the pipeline above ground, supported by permafrost, which engendered an innovative and creative set of solutions. An interesting major concern was to find a way not to interfere with the annual caribou migration. Security was (and is) a big issue. Some unanticipated risks also arose, some with unintended consequences. This paper examines the responses to myriad challenges, examining it from a systems engineering and systems thinking viewpoint. Questions for discussion are suggested so that this can be used as a case study in a course on systems engineering or systems thinking.

To examine the Project Management aspects of the French Air Defense Program SCCOA, a Model-Based System Engineering approach using the NATO Architecture Framework (NAF) is appropriate to ensure the System of Systems consistency. Two limitations of the NAF are addressed: incorporating temporality and incorporating decision support tools. The first issue is resolved by coupling NAF with an Access calendar database. The second is solved using Prolog, a Constraint Programming tool, and Cplex, a Mathematical Programming tool. The resulting tool stack allows to schedule deployment integrating Robust Optimization techniques.

For almost a year, the Design Studio and systems engineering teams at Dassault Systèmes have shared their respective practice: design thinking and complex systems engineering. This comparison gave us insights about several shifts: the people involved in project ecosystems, the call for more disruptive innovation, the growing capabilities of computers, the need to take into account the full complexity of humans and a few shared ambitions between both disciplines. After explaining this context, this paper reports on the comparison between the two practices, through a cross-referenced strength and weakness comparison, and other counterbalancing points. We also share early hypotheses, gleaned from our experiments, on how to combine the design thinking and systems engineering approaches in early stages of innovation, at the right time, despite cultural differences. To conclude, we look at what is needed to make complexity easier to grasp, how a combined approach also calls for a fresh look at project organisations and for a practice mixing art and technology.

We describe a validation approach for Simulink models of industrial cyber-physical systems (CPS), based on an adaptation of a coverage-guided test generation method for hybrid systems. Modelling an industrial CPS requires integrating heterogeneous components, which introduces high complexity in model verification. Using Simulink, which has become a de-facto industrial tool, heterogeneity comes from combining different formalisms (Simulink blocks, Stateflow diagrams, Matlab and C functions, etc.) and mixing different types of dynamics (discrete, continuous). Since the interactions between such components are often too complex to be faithfully captured in an existing mathematical modelling paradigm, we resort to treating them as black box systems while trying to exploit as much as possible a-priori knowledge about them. We first describe our approach: extracting from a Simulink model the information to define the main ingredients of the test generation framework, in particular environment inputs in which faults could be injected and critical states that require good coverage. We then illustrate the approach with an industrial model of an HVAC (Heating, Ventilation and Air Conditioning) system.

This paper presents an innovative approach for the engineering of complex SCPSs and large scale SoSs. It is based on the modelling and co-simulation of the dynamic phenomena determined by the different disciplines involved in the engineering of such systems. This covers the complete system lifecycle, from prospective studies aiming at defining the nature and scope of a system to be developed, down to system operation and maintenance, retrofits and modification.

In the past years, energy demand has increased and shifted especially towards electricity as the form of consuming energy. As the number of electric devices constantly grows and energy production must increasingly rely on renewable sources, this leads into noteworthy disparity between electricity production and consumption. This paper describes the results of a joint work between partners in ITEA2 12004 Smart Energy Aware Systems (SEAS) project, which aims towards providing the ICT tools and systems in order to help energy actors better manage and optimize energy consumption, production and storage. This paper presents and studies the innovative IT architecture proposed during this project, SEAS Reference Architecture Model (S-RAM). This architecture relies on four distributed services that enable to interconnect any energy actors and give them the opportunity to provide new energy services. The benefits of S-RAM have been studied on a specific use case, which aims to provide a service for estimating local photovoltaic production. It particularly helps energy management systems better plan electric consumption.

Specification of complex systems is a set of large documents written in natural language. Due to their complexity, they are often hard to understand and even harder to maintain. We designed the domain specific language ScOLa (Scenario Oriented Language) to model the architecture and behavior of systems using a set of formalized concepts in order to support the dialog between experts. In this article, we present a reverse engineering methodology to formalize complex system specifications using scenarios. It starts from an informal description of the system and results in a hierarchical view of the system description. This article aims both at introducing ScOLa and at presenting its application on the railway systems.

Combining subsystems to build a fully integrated product is a challenging task in complex systems design. The integration of flow components requires a fast creation and validation of different pipe route variants. In this article an algorithm for the automated generation of pipe routes in a given installation space is presented. The pipe route generation is constrained to the usage of prechosen (standardized) pipe bend sets. The routes are rule-based manipulated and evolved using a simulated annealing optimization scheme.

As the supply of desalinated water becomes significant in many countries, the reliable long-term operation of desalination infrastructure becomes paramount. As it is not realistic to build desalination systems with components that never fail, instead the system should be designed with more resilience. To answer the question how resilient the system should be, we present in this paper a quantitative approach to measure system resilience using semi-Markov models. This approach allows to probabilistically represent the resilience of a desalination system, considering the functional or failed states of its components, as well as the probability of failure and repair rates. As the desalination plants are connected with the end-user through water transportation and distribution networks, this approach also enables an evaluation of various network configurations and resilience strategies. A case study addressing a segment of the water system in Saudi Arabia is given with the results, benefits, and limitations of the technique discussed.

Multi-objective evaluation is a necessary aspect when managing complex systems, as the intrinsic complexity of a system is generally closely linked to the potential number of optimization objectives. However, an evaluation makes no sense without its robustness being given (in the sense of its reliability). Statistical robustness computation methods are highly dependent of underlying statistical models. We propose a formulation of a model-independent framework in the case of integrated aggregated indicators (multi-attribute evaluation), that allows to define a relative measure of robustness taking into account data structure and indicator values. We implement and apply it to a synthetic case of urban systems based on Paris districts geography, and to real data for evaluation of income segregation for Greater Paris metropolitan area. First numerical results show the potentialities of this new method. Furthermore, its relative independence to system type and model may position it as an alternative to classical statistical robustness methods.

Quantitative and objective management of complexity is essential for effective design of engineered complex systems. In this paper, we develop a quantitative framework for complexity management. This includes a measure of system value that explicitly considers system complexity. The system design goal is to maximize the system value. Using a simple, representative mathematical model linking performance to system complexity, we show analytically that there exists a regime where we have an optimal level of complexity that leads to maximization of system value. Existence of this regime is dependent on two rate parameters that link the complexity-performance-development cost triad for engineered systems. Outside of this regime one has to always aim for reducing system complexity in order to maximize system value. The framework is subsequently applied to a case study involving a set of aircraft engine architectures.

This paper presents the results of human subject experiments focusing on the role of decision rules in the study of flexibility and real options analysis (ROA) in design and management of complex engineering systems. Decision rules are heuristics-based triggering mechanisms that help determine the ideal conditions for exercising flexibility in system operations. In contrast to standard ROA based on dynamic programming, decision rules can be parameterized as decision variables, and therefore capture the decision-making process based on specific realizations of the main uncertainty drivers affecting system performance. Similar to standard ROA, a decision rule approach can be used to quantify the benefits of flexibility in early conceptual design studies, and help identifying the best flexible systems design concepts before a more detailed design phase. While many studies demonstrate expected lifecycle performance improvement stemming from a decision-rule based approach as compared to standard design and ROA techniques, very few studies show experimentally their effectiveness in managing flexible engineering systems. This paper presents the results of controlled human-subject experiments involving thirty-two participants evaluating a training procedure in a simulation game environment. The controlled study show that a stochastically optimal flexible strategy combined with an initial policy for the system configuration can improve significantly the expected coverage rate of medical emergencies. These provide insights for further research, development and evaluation of flexible systems design and management strategies for complex engineering systems.

This case study analyses the applicability of a Quality Improvement Process that will enhance the quality of the requirements using an existing requirement specification to seed the Knowledge Base with the organization know–how. This Knowledge Base drives the quality assessment and directs the requirement authors to the areas that can be improved. The updated specification feeds back into the Knowledge Base thereby institutionalizing continuous process improvement into Alstom. The case study has been developed by means of a Proof of Concepts using the RQS suit tools to gather the knowledge (KM), analyse the quality (RQA) and authoring requirements (RAT).

The goal of this paper is to explore the need for Systems Engineering (SE) in East African countries and how best to educate engineers in the field of SE. It provides a comprehensive overview on the usefulness of SE education for East African nations and proposes SE curriculum. Presently SE has been given little attention in East African countries. However, these countries are in the beginning of industrialization with many new mega projects and infrastructure expansions that demands SE professionals. This motivates the need to introduce SE education in the region. Systems engineering education demands the development of SE curriculum, which will be multidisciplinary and considers social and psychological factors by taking into account the active participation of the community. Towards this end, the paper demonstrates the necessity of SE education via designing and managing ongoing mega projects. It also proposes SE curriculum by incorporating courses covering both foundations of SE and the practice of SE. Furthermore, it suggests that these courses will be delivered in collaboration with industry and government entities.

This paper presents the challenges in ensuring the existence of a workforce this is capable of conceiving, realizing and supporting increasingly complex systems throughout the lifecycle. Accomplishing this will require that all systems decision makers are systems thinkers, all engineers have systems engineering skills, and all systems engineers are broad-based technical leaders. The human capital development and academic forum research focus areas of the Systems Engineering Research Center (SERC) and the International Council on Systems Engineers (INCOSE), respectively, are presented in this work. In addition, current status and future efforts are discussed.

System engineering education is major challenge with regard to the current structure and organization of the higher education system oriented towards specialization. Although several programs have emerged in the recent years in France in universities and school of engineers the field remains broadly unknown. The number of young graduates with system engineering education remains low. At the same time, system engineering is increasingly used in industry now beyond the usual defense and aeronautical applications. As a result continuous education and training needs for confirmed engineers have surged in the recent years. In this paper we describe a feedback on the FAIS program for confirmed engineers with an analysis of the various parameters which have clearly contributed to its success. Several trends have emerged as a result of this program. System engineering is in need of theoretical foundations and this is increasingly expressed by system engineering participants. Requirements engineering and architecture were initially major themes of the training with the use of associated softwares (DOORS, IBM System Architect, MEGA). Significant experience have been achieved in coaching trainees in understanding and mastering both the concepts and practice on significant case studies. However, the trend have increasingly be in expectations from trainees for architecture evaluation and mastering architecture complexity. The FAIS program have strongly evolved during the past 6 years and have better adapted to the growing needs of attendees. A turning point have been the adoption of coaching and supervision of attendees on their own case studies. Departing from a common academic style case study to the current projects of DGA attendees have both enriched the training and increased the involvement of attendees in their training. A trusted environment of learning and exchange have also allowed deepening of the system engineering issues raised by the case studies. This have been allowed by the common work origin of the participants. MBSE, architecture modelling and evaluation have increasingly dominated the training and the trend is confirmed in 2016. Continuous improvements have been brought with growing training hours for some topics and diminishing training hours for others. FAIS have operated as a sliding window on education topics. Multidisciplinary system simulation, mathematical modelling and optimization coupled with simulation are the emerging trends of the program.

To achieve the full potential of PLM in Systems Engineering tools especially in view of the system’s complexity in industries such as the consumer industry a clear understanding of how best to use such systems is important to product development activities. Systems Engineering is an interdisciplinary field of engineering that focuses on how to design and manage complex engineering systems over their life cycles. Issues such as reliability, logistics, coordination of different teams (requirements management), evaluation measurements, and other disciplines become more difficult when dealing with large or complex projects. Systems Engineering deals with work-processes, optimization methods and tools in such projects. It overlaps technical and human-centered disciplines such as control engineering, industrial engineering, organizational studies, and project management. Systems Engineering ensures that all likely aspects of a project or system are considered, and integrated into a whole. After a short introduction, this paper, which is based on the results of the accomplished descriptive study and literature survey of the Design Research Methodology according to Blessing and Chakrabarti, presents a generic integrated approach of System Driven Product Development (SDPD) and demonstrates the general requirements of a generic integrated approach during the Engineering Design of Systems. The second section presents a new approach of Systems Engineering, which is based on SDPD and will explain the different phases and sub-phases of the developed approach. By means of designing an electric skateboard the different phases of the developed generic integrated approach will be demonstrated and presented. Section three will discuss the results of the Prescriptive Study and address the most important issues. In general, this paper presents the Prescriptive Phase of the Design Research Methodology.

The increasing complexity of software systems is constantly fueling the interest in pragmatic analysis methods. These are by no means scarce, but their applicability requires additional expertise that often has a weak relation with the development process or the domain the system is intended for. The model-driven paradigm addresses this issue at a certain extent by raising the level of abstraction closer to the domain and facilitating development and analysis by means of automation. It tries to shift the inherent complexity from the model towards the automation process. Although this has proven to be quite effective in handling functional aspects, the same cannot be stated with confidence regarding non-functional aspects like performance. In this paper we present a model-driven approach for performance analysis based on standardized languages. The functional aspects of the system are captured using SDL and enriched with performance annotations. Available resources are assigned to system components via deployment diagrams, and real test cases described in TTCN-3 drive model execution. Different scenarios can be executed automatically, and the graphical presentation of results can aid the user to decide on the best allocation of resources in terms of execution time and payload.

Network models have already been used with the intent to gain additional information about the structure of product development processes (PDP). These are supposed to map the flow of information and data as well as to provide a deeper understanding of the company’s procedures. Process networks commonly represent dependencies of tasks and/or social contacts. Treating tasks as nodes in a network allows for a comparison of their position within the process. This way, it is possible to characterize certain actions according to their network attributes. In order to fully describe a PDP, it is, however, necessary to include other influencing factors as well. For example, there are only few approaches examining the impact of quality and progress on the process artefacts (such as CAD-files). The goal of this paper is to clarify what information is necessary to precisely describe a PDP in a network model. This covers a statement about the level of detail, general structure and dynamic of the networks.

Agile development as an alternative to traditional plan-driven approaches gains rising popularity in both software and non-software industries due to its advantages in dynamic and uncertain environments. Although its implementation challenges are widely explored, interdependencies between them are mostly neglected in recent papers. Practice and academia, therefore, often try to find local optimizations without (a) considering the interdependencies and (b) differentiating between causes and effects. By using the network theory this investigation sets up a directed network containing 241 challenges (nodes) and 360 dependencies (edges) and executes a cause-and-effect analysis. To identify challenges that are most crucial and, thus, of highest importance for future research, the analysis takes each challenges’ (i) degree of being a cause, (ii) impact and (iii) range of influence into account. ‘Granting freedom of action and decision’, ‘integrating agile methods in traditional organizations’ and ‘composing agile teams’ turn out to be the top three challenges.

The development of critical systems is a challenging task that requires collaborative work for various purposes: specification, design and verification. Today, no single modeling language and environment covers all these aspects. ARCADIA and Capella^© are Model-based System Engineering (MBSE) method and tool developed for the system design process. ARCADIA/Capella also adopts a viewpoint-based description to describe engineering specialty, such as the safety engineering. Safety Architect^© is a MBSA (Model Based Safety Analysis) tool developed by ALL4TEC to analyze the robustness of design models. Indeed, Safety Architect can use design models imported from usual modelling tools, such as Capella, in order to perform classical safety analyses: automatic deduction of fault tree of the identified feared events. In this paper, we present our MBSE and MBSA approach developed in the Clarity project around Safety viewpoint in Capella and the import legacy into Safety Architect in order to realize safety analysis.

The analysis of infrastructure network reliability is an important task required for disruption prevention, protection or recovery planning. In order to truly encapsulate the actual structure of infrastructure networks, it is proposed to analyze the infrastructure networks with heterogeneous nodes, i.e. the nodes with distinct operating features in the network. In this paper, an infrastructure network with heterogeneous nodes is modeled as a graph with a set of nodes with supply feature, a set of nodes with demand feature, and a set of connections between the nodes. The network resilience can then be evaluated by the weighted sum of all the resilience of the demand nodes, so the proposed resilience analysis approach can be used to indicate the ability of the network to resist disruption.

We analyze the “smartness” of the city as based on an organic evolving system. The smart city, in this perspective, is not a first order cybernetic self-regulating system which the number of parameters and variations could be finite which culd be modeled top- down by an engineer. On the contrary, the smart city is an autopoietic ecosystem (Maturana) and an adaptive system as promoted by the second order cybernetics, able to evolve as a dissipative system thanks to its internal interactions faced with the variations of its environment.

Areva investigated a Model Based approach for setting up the Work Breakdown Structure of a nuclear plant project performed in collaboration with other key industrial partners of the energy domain.

arKItect SEA, a Systems Engineering (SE) modeler has been used by Renault and Bouygues Energies & Services to manage SE processes of a Batteries for Buildings (B4B) system. B4B is a concept reusing batteries of electric vehicles (second life), as a storage facility for energy management and renewables integration. The project started in 2012, was completely new from many viewpoints: new partnership with actors using different processes; innovative product and service offer connected and evolutive including safety concerns.

Henderson and Clarck (1990) have introduced the sole framework which classifies technical change in a system using two measures: degree of changes in components and intensity of changes in the linkage between components. Although this two dimensional framework is useful for understanding the congruence between different kinds of technical change, their consequences for the system’s performance and their required capabilities, it ignores the vastness and relative importance of changes. To cope with this challenge, we propose adding a third dimension entitled “Change Magnitude” to their framework which contains a spectrum from changes in just one peripheral component or linkage to changes in all peripheral as well as core components and linkages. The resolved framework, presents an octal categorization of technical change in systems which provides a better basis for classification.

Cost analysis is challenging for multiple reasons, one of which is the lack of historical data due to proprietary issues, or significant work required to make it useful for a particular application and domain of interest. In addition, to support system engineering methods such as Design Space Exploration, both component- and engine-level costs must be supported. This paper presents the results of a preliminary study on a tool that can be used to estimate the development cost for a set of airplane-engine architecture models using publicly available off-the-shelf tools. Our tool focuses on supporting complex system engineering tool chains and methods that require strong interoperability with different tools in a networked environment. The tool, through its architecture, allows the inclusion of supports for early stage cost analysis without directly using historical data, and both system- and component-level cost generations. We describe our approach, tools, estimation process and possible use cases.

A regional production network could be viewed as a complex network, consisting of an intertwined set of supply chains in a bounded geographical space, linking multiple customers in a particular industry with their associated suppliers. To avoid suboptimal decisions, supply chain managers and policy makers need to recognise the structural complexity of the regional production networks in which the individual supply chains are embedded. The authors propose a framework that allows for the identification of complexity traits in the regional production network structure, which provides an insight into its functionality and operational characteristics. The framework is based on the identification of network topology and structural parameters, including density, clustering and average path length. These parameters are indicative of network responsiveness, adaptability and resilience. The authors have applied the proposed framework to empirical data from the South Australian resource extraction sector to highlight how the regional production network structure could be used as a dashboard to assist both practitioners and policy makers in supply chain governance decision making.