Integrated design and operation management for enterprise ... 2010... · W.J. Zhang, J.W. Wang & Yingzi Lin To cite this article: W.J. Zhang, J.W. Wang & Yingzi Lin (2019) Integrated

Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=teis20

Enterprise Information Systems

ISSN: 1751-7575 (Print) 1751-7583 (Online) Journal homepage: https://www.tandfonline.com/loi/teis20

Integrated design and operation management forenterprise systems

W.J. Zhang, J.W. Wang & Yingzi Lin

To cite this article: W.J. Zhang, J.W. Wang & Yingzi Lin (2019) Integrated design and operationmanagement for enterprise systems, Enterprise Information Systems, 13:4, 424-429, DOI:10.1080/17517575.2019.1597169

To link to this article: https://doi.org/10.1080/17517575.2019.1597169

Published online: 30 Mar 2019.

Submit your article to this journal

Article views: 449

View related articles

View Crossmark data

Citing articles: 2 View citing articles

EDITORIAL

Integrated design and operation management for enterprisesystems

Enterprise is a mini social-technical-ecological system in that it consists of humans, equip-ment and machines, and it has a location or site. Its structure follows the substance-infrastructure (S-I) framework (Zhang and Wang 2016; Zhang and van Luttervelt 2011).There are two types of the S-I framework: the substance drives the infrastructure (Type I) andthe infrastructure drives the substance (Type II). The enterprise system belongs to Type II.For instance, to a manufacturing system, the substance refers to goods made of materials,and the infrastructure refers to humans and machines, which produce and deliver goods tocustomers in response to their demands. To a service system (Wang et al. 2014), thesubstance refers to data (knowledge and information) (Zhang 1994) or signals or humans,and the infrastructure refers to humans and machines, which generate data, producesignals, or offer services to customers in accordance with their demands.

Enterprise is a dynamic system, and it changes in its state and/or structure withrespect to time, location, and/or event, and both the substance and infrastructure maychange. A change on the part of the structure and/or state, say A, always has a reason orreasons, and this change is further associated with the change of another part of thestructure and/or state of the system, say B; B is an independent variable and A isa dependent variable in this case (Zhang et al. 2005). The knowledge that governs therelation of A and B is called principle (Zhang et al. 2005; Zhang and Wang 2016). Forinstance, B is the force (F) applied on a block and A is the acceleration (a) of the blocksystem, and the knowledge that governs the relation of A and B, in this case, is theNewton’s second law, that is, F = ma, where m is the mass of the block system. A caremust be taken that the principle (knowledge) may be hidden or unfolded in data or bigdata but a correspondence relation of A and B can be built using various machinelearning methods, e.g., Artificial Neural Network (ANN) (Zhao and Zhang 2017), variousdeep learning methods (Zhang et al. 2018), etc. The independent variable is a functionof time, location, and/or event, so is the dependable variable, and thus the whole systemchanges with respect to time, location, and/or event.

Design of an enterprise system means to determine its structure (infrastructure andsubstance) in response to a need or demand in a context (Zhang and Wang 2016). Forinstance, in response to the need of charging to electric vehicles, a new enterprise idea,the electric charge station enterprise, emerges. To this new enterprise, one needs todetermine the charging equipment, number of workers, and so on, which makes senseto the design of an enterprise (Zhang and Wang 2016). Construction of an enterprisefollows its design. Design and construction are processes, so it makes sense to say abouttheir management. A good practice of the management of design and construction thusresults in a good structure.

ENTERPRISE INFORMATION SYSTEMS2019, VOL. 13, NO. 4, 424–429https://doi.org/10.1080/17517575.2019.1597169

© 2019 Informa UK Limited, trading as Taylor & Francis Group

After an enterprise is designed and constructed (i.e., its structure is available), theoperation of the enterprise takes shape. The operation includes: planning, scheduling,and executing. It makes sense to say about their management called operations manage-ment. A good practice of the management of these operations, together with the structureof the enterprise system, eventually results in a cost-effective performance (i.e., supplymeeting demand) along its behavioural properties (i.e., stability, reliability, robustness,resilience, and sustainability) of the enterprise system. It is to be noted that to a roboticsystem, the operation activities (i.e., planning, scheduling and executing) are crashed intoa compact one called ‘actuation’, and management changes to control accordingly.

Figure 1 shows the relationship among design, operation, structure (including customerand technical specification), and management. By generalisation, management hasa target to be managed, such as design and operation activities. Management is tocontrol the running of the activities to achieve a certain goal under a certain constraint(Figure 1). The goal of design management is to generate a specific structure of a systemalong with customer voice of needs and technical specification (Figure 1), and the goal ofoperation management is to make the structure to a specific performance along with a setof specific properties (e.g., stability) that the structure is expected to exhibit (Figure 1). Dueto the presence of imprecise structure and data (Cai et al. 2017), the feedback approach tomanagement is widely employed (Figure 1; Loop 1 – Loop 4). The feedback managementmay also take place on Loop 5 (Figure 1), which involves design activities and operationactivities and their managers. In feedback management, management is a decision-making process, so it is a system called management system or manager.

The manager is a cognitive system rather than a physical system. As such, the structureof a management system is essentially a symbolic expression, e.g., y = f(x), where f(.) isa mapping, and x and y represent objects in the domain of concepts which are structure,state, principle, behaviour, context, and function according to the FCBPSS architecture(Zhang et al. 2005; Zhang and Wang 2016). A different f means a different managementsystem or manager. A manager could be a human being or computer (or machine). Toa human manager, f is hardwired in the human brain, and to a computer manager, f iscoded. In a particular management process, f(.) may not be unchanged. A manager(design manager, operation manager) that its f(.) may change during the management

CTS Performance along with property Design &

Construct Operation

Design

Manager

Operation

Manager

Operation

Database

Design

Database

1 2 3

4

5

Meta

Design

Manager

Meta

Operation

Manager

Information flow Act on Process Data

Figure 1. Relationship among design, operation, and management. CTS: Customer- Technical-specification-Structure.

ENTERPRISE INFORMATION SYSTEMS 425

process is called adaptive manager, and the change is taken by meta manager (meta-design manager, meta operation manager), see Figure 1. The meta-design manager (metaoperation manager) is mapping as well, denoted by g(.), and δf(.) = g(z), wherez represents the structure (performance), and δf(.) represents the modification f(.).

It is noted that when the design manager (fd) changes, the structure of the system(e.g., enterprise) under design and construction changes. If a structure has already beenbuilt, the change of the structure requires that the structure is designed as a changeableor adaptable structure. Further, the implementation of the change may be performed bythe structure itself (self-change) or by the external entity.

Design management and operation management (design and operation in short) inthe above must be integrated because both refer to the same entity (the general reasonfor integration). This integration theory for systems may be called IDOM (IntegratedDesign and Operation Management), which is an extension of the integrated design andcontrol for robotic systems (Li, Zhang, and Chen 2001, Cheng et al. 2012). In thefollowing, several axioms for IDOM are proposed by generalising the existing resultsfrom integrated design and control of robots.

First, a robot may be modularised, and a module may include several parameterswhich further correspond to multiple performance indexes. As such, by selection of onemodule, all the parameters of this module are selected, which means that all theperformances related to these parameters must be considered together (Bi and Zhang2001), which gives a sense of integration. By a generalisation of this scenario, anintegration axiom (IDOM Axiom I) can be concluded as follows:

IDOM Axiom I: When a module or a set contains a number of parameters (pi, i = 1,2, . . ., n), which correspond to a number of performances or properties (ppj, j = 1, 2, . . ., m),the simultaneous determination of pi with the consideration of all the ppj under theconstraint that the modules are available, say mk (k = 1, 2, . . ., w), and each module ischaracterized by pi is of necessity. This integration problem can be mathematically repre-sented as a constrained multi-objective optimization problem with discrete variables.

Second, the simultaneous determination of the parameters, particularly one groupthat describes the structure and the other group that describes the operation, may leadto a better performance than the sequential determination of the parameters in robotics(Zhang and Chen 2000; Cheng et al. 2012). By a generalisation of this scenario, one canlead to another integration axiom (IDOM Axiom II) as follows:

IDOM Axiom II: If a set of parameters across two or more phases (e.g., the parameterset A in phase I, and the parameter set B in phase II) are coupled to one or moreperformances, the simultaneous determination of these parameters for the optimal perfor-mances may lead to improved performances as opposed to the sequential determination ofthe parameters based on the precedence of the phases (i.e., the phase I precedes the phaseII; A in the phase I is determined first, followed by B in the phase II).

Finally, in robotics, a well-known approach to the integration of design and control isto design the structure of a robot such that some properties of the structure can beobtained, e.g., cancellation of gravitational force of the robot, which reduces the‘burden’ of the controller and subsequently improves the overall performance of therobot further. This approach in robotics is called DFC (Design For Control) (Zhang, Li,and Guo 1999). The philosophy behind DFC can be explored. One can view control as anintervention to a system. To each burden, a particular control is designed. Suppose that

426 EDITORIAL

the system has several burdens say A, B, C. The control for A may have bad side effectson the control for B and C (respectively), and the control for B may have bad side effectson the control for A and C. So if the structure can be designed to remove some burdensto the control, which will increase the chance of minimising bad side effects across thedifferent controls and thus improve the overall performance of the robot. Bya generalisation of this scenario, one can lead to another axiom for IDOM (IDOMAxiom III) as follows:

IDOM Axiom III: Suppose the parameter set A is in the phase I and the parameter setB is in the phase II and the phase I precedes the phase II. Identify the potential burdensto the phase II, and perform the activities in the phase I such that some of the burdenscan be removed.

Several remarks can be made regarding the above IDOM approach. Remark 1: Anyintegration scheme that goes beyond the above axioms may actually get some badresults, as the integration can inherently create unnecessary constraints, which may losesome promising solutions called bad integration. Remark 2: In case that the phase I isdesign and the phase II is operation, IDOM can only be performed based on thesimulation unless the structure can be changeable or adaptable, simulation of theactivities on the phase II in particular. Remark 3: The three axioms of IDOM areapplicable to activities in one phase (e.g., the planning and scheduling activities in theoperation phase) or to activities across different phases (e.g., the structure in the designphase and the plan and schedule in the operation phase). Remark 4: An enterprisecould be a single organisation or a group of organisations, especially a chain oforganisations (Figure 2). In the latter case, the IDOM approach is applicable to thegroup with (1) integration on activities in one phase (Integration 3 in Figure 2), (2)integration on activities across different phases (Integration 2 in Figure 2), and (3)integration on activities across different organisations, e.g., Structure 1 and Structure 2(Integration 1 in Figure 2). Remark 5: The integration problem can be formulated as anoptimisation problem model, and it is a non-trivial issue to solve such a model effec-tively and efficiently. Remark 6: Integration also makes sense to integrated design ofsubstance (products) and of infrastructure (manufacturing systems), a specialisation ofwhich is the so-called design for manufacturing approach, as well as integrated designof operation and of operator or manager.

This special issue collected seven papers on the theme of integration in various ways yetunder the framework as illustrated in Figures 1 and 2. Paper 1 entitled ‘A novel resilientscheduling paradigm integrating operation and design for manufacturing systems withuncertainties’ proposes an approach to integrated system design and operation schedulingfor the task-specific production performance along with the resilience property. The

Figure 2. Integration schemes on two organisations. D: Design, C: Construct, O: Operation, P:performance or property, CTS: Customer-Technical-specification-Structure.

ENTERPRISE INFORMATION SYSTEMS 427

integration scheme of the paper hits Integration 2 in Figure 2. The problem is further solvedby a method that incorporates big data analytics into a multi-stage optimisation method.Paper 2 entitled ‘An integrated approach for dynamic customer requirement capturing forproduct development’ proposes an integrated approach to capture the customer’s voice ofneeds and technical specification, which are two activities in the design phase for productsor enterprises that make the products. The integration scheme of the paper hits Integration4 in Figure 2. Paper 3 entitled ‘Aspect-oriented challenges in system integrationwith micro-services, SOA and IoT’ discusses tools with which integration can be enabled, AOP (Aspect-Oriented Programming) in particular. A promising integration can only be realised upon anaccurate understanding of each system and their interactions, and not all integrations aregood according to Remark 1 in the above discussion. The contribution of this paper is theprovision of information about AOP and its role in integration. Paper 4 entitled ‘Integratedscheduling of production and distribution operations in a global MTO supply chain’ dis-cusses the problem of the integration of production scheduling of products and theirdistribution in a global MTO supply chain with two objectives: cost and delivery time. Theintegration scheme of the paper hits Integration 3 in Figure 2. The contribution of the paperis a new formulation of the problem and a new solution to it. Paper 5 entitled ‘Patientassignment scheduling in a cloud healthcare system based on Petri Net and greedy-basedheuristic algorithm’ proposes an approach to integrated resource planning and patientassignment scheduling for a group of hospitals, which share the same resource. Theintegration scheme of the paper hits Integration 5 in Figure 2. Paper 6 entitled ‘Themodelling and operations for the digital twin in the context of manufacturing’ developsa platform for digital modelling and simulation of products, manufacturing or constructionprocesses, operations or actions. The platform is a tool for integrated design and construc-tion or building or manufacturing of products, i.e., Integration 4 in Figure 2. Paper 7 entitled‘integrated scheduling for a distributed manufacturing system: a stochastic multi-objectivemodel’ develops a model for integrated scheduling among a group of firms with considera-tion of uncertainties. The integration of the paper hits Integration 5 in Figure 2.

The IDOM is still under development, and its potential to enterprise system designand operation management is still being explored. The main issues for future IDOM are:(1) how to efficiently solve the IDOM problem model, as it is usually a multi-objectiveoptimisation model and the computational challenge is high, (2) how to identify anyunnecessary integration, as integration of two unrelated activities is creating unneces-sary constraints that degrade their performance, (3) how to optimise the integrationscheme for complex network and distributed systems such as holistic supply chainnetwork systems to improve their task performance as well as their robustness andresilience (Wang et al. 2016, 2017; Said, Bouloiz, and Gallab 2019).

References

Bi, Z. M., and W. J. Zhang. 2001. “Concurrent Optimal Design of Modular Robotic Configuration.”Journal of Robotic Systems 18 (2): 77–88. February. doi:10.1002/1097-4563(200102)18:2<77::AID-ROB1007>3.0.CO;2-A.

Cai, M. Y., Y. Lin, B. Han, C. Liu, and W. Zhang. 2017. “On a Simple and Efficient Approach toProbability Distribution Function Aggregation.” IEEE Transactions on Systems, Man, andCybernetics: Systems 47 (9): 2444–2453. Sept. doi:10.1109/TSMC.2016.2531647.

428 EDITORIAL

Cheng, L., Y. Lin, Z.-G. Hou, M. Tan, J. Huang, and W. J. Zhang. 2012. “Integrated Design of MachineBody and Control Algorithm for Improving the Robustness of a Closed-Chain Five-Bar Machine.”IEEE/ASME Transactions on Mechatonics 17 (3, JUNE): 587–591. doi:10.1109/TMECH.2012.2183378.

Li, Q., W. J. Zhang, and L. Chen. 2001. “Design for Control (DFC): A Concurrent EngineeringApproach for Mechatronic System Design.” IEEE/ASME Transactions on Mechatronics 6 (2):161–169. June. doi:10.1109/3516.928731.

Said, S., H. Bouloiz, and M. Gallab. 2019. “A New Structure of Sociotechnical System ProcessesUsing Resilience Engineering.” International Journal of Engineering Business Management.doi:10.1177/1847979019827151.

Wang, J. W., H. F. Wang, W. J. Zhang, W. H. Ip, and K. Furuta. 2014. “On a Unified Definition of theService System: What Is Its Identity?” IEEE Systems Journal 8 (3): 821–826. doi:10.1109/JSYST.2013.2260623.

Wang, J. W., R. L. Dou, R. R. Muddada, and W. J. Zhang. 2017. “Management of a Holistic SupplyChain Network for Proactive Resilience: Theory and Case Study.” Computers & IndustrialEngineering. doi:10.1016/j.cie.2017.12.021.

Wang, J. W., R. R. Muddada, H. F. Wang, J. L. Ding, Y. Lin, and W. J. Zhang. 2016. “Towardsa Resilient Holistic Supply Chain Network System: Concept, Review and Future Direction.” IEEESystems Journal 10 (2): 410–421. doi:10.1109/JSYST.2014.2363161.

Zhang, W. J. 1994. “An Integrated Environment for CADCAM of Mechanical Systems.” Ph.D. thesis., DelftUniversity of Technology, ISBN 90-370-0113-0, pp. 1–263. doi:10.3168/jds.S0022-0302(94)77044-2.

Zhang, W. J., and C. A. van Luttervelt. 2011. “Toward a Resilient Manufacturing System.” CIRPAnnals – Manufacturing Technology 60: 469–472. doi:10.1016/j.cirp.2011.03.041.

Zhang, W. J., G. Yang, Y. Lin, C. Ji, and M. M. Gupta. 2018. On Definition of Deep Learning, 2018World Automation Congress (WAC). Stevenson, WA. 2018. 1–5. doi:10.23919/WAC.2018.8430387.

Zhang, W. J., and J. W. Wang. 2016. “Design Theory and Methodology for Enterprise Systems.”Enterprise Information Systems 10 (3): 245–248. doi:10.1080/17517575.2015.1080860.

Zhang, W. J., Q. Li, and S. L. Guo. 1999. “Integrated Design of Mechanical Structure and ControlAlgorithm for a Programmable Four-Bar Linkage.” IEEE/ASME Transactions on Mechatronics 4 (4):354–362. Dec. doi:10.1109/3516.809514.

Zhang, W. J., and X. B. Chen. 2000. “Mechatronics Design for a Programmable Four-BarMechanism.” Proceedings of I. Mech. E. Part C, Journal of Mechanical Engineering Science 215:365–375. doi:10.1243/0954406011520670.

Zhang, WJ, Y. Lin, and N. Sinha. 2005, “On The Function-behavior-structure Model for Design,”Proceedings of the Canadian Design Engineering Network (CDEN), 2nd InternationalConference, Kaninaskis, Alberta, July 18–20.

Zhao, Y., and W. J. Zhang. 2017. “Developing a Mapping from Affective Words to DesignParameters for Apparel Design by Applying Artificial Neural Network.” Textile Research Journal.doi:10.1177/0040517516669072.

W.J. ZhangSchool of Mechanical Engineering, Donghua University, Shanghai, China,

Department of Mechanical Engineering, University of Saskatchewan, [email protected]

J.W. WangDepartment of Industrial and Manufacturing Systems Engineering, University of

Hong Konghttp://orcid.org/0000-0001-8895-2214

Yingzi LinDepartment of Mechanical and Industrial Engineering, Northeastern University,

Boston, USA

ENTERPRISE INFORMATION SYSTEMS 429