Decision making support in supply chains through modified petri nets and case-based reasoning
5 October, 2016
Atlantis University Admissions
11 October, 2016

An Information System to Support the Decision Making Process in Supply Chain Management

Dr. Inty Saez Mosquera, Inty.Saez@atlantisuniversity
Dr. Jorge Marx Gómez,
Dr. Gilberto Hernández Pérez,

From the organizational point of view, one of the main problems in the administration of the Supply Chains is the strategic alignment of all the members of the chain. This difficulty, in most of the cases, is determined by the ability of the partners to share strategies decisions in a logistical management environment.

Independently of the context of the decisional problems that the members face, in context of the logistical management of the chain, it is possible to identify common patterns. Generally, the persons in charge of making the decisions use their “memories” of their behavior in previous situations to create a new behavior that is tailored to the new conditions. We propose a general approach to modeling problem formulation and solution for theses problem to identify the common patterns previously mentioned, integrating the knowledge resulting and share it through Supply Chain.

Keywords: SCM, decision-making alignment, Petri Net, CBR, knowledge sharing and integration


From the organizational point of view, one of the main problems in Supply Chains management is the strategic alignment of all the members of the chain [1]. This difficulty, in most of the cases, is determined by the ability of the partners to share strategies decisions in a logistical management environment [2,3].

Independently of the context of the decisional problems that the members face, in context of the logistical administration of the chain, it is possible to identify the common patterns [4,5]. These common patterns, that are shared, would constitute the base of a corporate benchmark running the length and width of the chain whose fundamental comparative indicator would be in fact, the effectiveness of the decisions made.

From this perspective, the problem of alignment of the strategic management of the organizations with its informative strategy seems to be an essential key element. However, is necessary to formulate the following question: what type of information is necessary to have available to reach these objectives? The use of the term of information is polisemic [6], in the context that it is being used, the patterns previously mentioned, constitute the nature and content of the information that should be shared among the members of the supply chain.

Let us take the Theory of the Limitations for example (TOC), and let us use the DBR to refer to a strategic decision in the context of logistical management and the handling of its capacities [2,7]. This solution has a known pattern and, depending on the operative conditions of the organization, it can be either less or more effective. In the same way, it is possible to extract the decisional pattern of the solutions based on MRP I, II, III, and so forth with the rest of the traditional managements approaches. These patterns, in each previously mentioned example, constitute the nucleus of the management solutions that promulgate each administration tools.

It is possible to keep up the thesis presented up to this point, based on the studies carried out by Nutt (1993, 1999) and presented by Corner [8] with a total of around 163 and 340 operative and strategic decisions respectively. With the emphasis on the verification of the fact that at the time of making the decision the persons in charged would search their memories for similar situations, learn and in each new interaction would incorporate this new knowledge. Generally, the persons in charge of making the decisions use their “memories” of their behavior in previous situations to create a new behavior that is tailored to the new conditions. Being in this way, the Case Based Reasoning (CBR) seems to be an alternative to support this behavior [9].

The ideas commented previously already have references in the scientific community. Biswas and Narahari [2] proposed the use of the paradigm from Object Oriented Modeling (OOM) to create a workbench for the modeling and evaluation of decisions in the supply chains context.


Information system is a well structured and studied research topic. Many contributions have been reached since years ago. In the context of decision-making support the promises of this system are far away to be reached yet [10,11,12,13,14]. They are many differences between support and aid in the context of decision-making process. Support means that the information to take decision exist and are available to person in charged of this process and after by using some procedure/technique or model, the decision will be taken.

Decision aid implies a commitment between the analyst (methodological knowledge, domain independent) and user (domain knowledge concerning the decision process). The final objective of this process is to arrive to a consensus between the user and the analyst [15]. The present state of information system describe a large number of unsuccessful use in many organization that invest a considerable amount of money in this system (such as ERP System [16,17,18,12,19,20]).


Figure 1. Relationship between information system and decision support efforts. Evolutionary perspective.

Even though, the information systems can be cover all enterprise information sources, managers don’t use all information in decision-making process. Managers use only a few views that have means for them in the context of their activities. The present approach for support decision-making process is based on data & information integration. Figure 1 illustrates the relation between gathering information for decision-making process and the acquisition & share knowledge.

We propose a new approach to support decision-making process in supply chain management. We propose modeling, store and sharing knowledge in this context. The present state of the investigation in artificial intelligence and knowledge modeling show that the opportunity to reproduce the way that human being think is far away at the moment. Furthermore, the human being thinking process and the ability to formulate and solve problem has many different aspects from one people to other.

Our approach is not interested in the way that human being creates problem & solutions formulations. Information in decision making context has the same role of raw material in manufacturing production environments. Based on information about situation and possible consequence of some actions, decision maker builds decision-making problem & solution formulation. Decision maker creates models in learning environments [21] and for that reason the approach suggest the creation of an easy way to express and interchange the knowledge that decision maker create/reuse in decision-making process. In this context, information means knowing (with some degree of reliance) the relationship between the manifestation of problem and the elements responsible for the non desired behavior of the system.


From Ausubel’s perspective [22] it is a significative learning when the process uses a common language for structuring all lecture and when the teacher uses a mechanism to make the student to uses a previous knowledge. Nutt (1993, 1999) found that managers use a learning approach to formulate and solve decision-making problem by using their previous memories about “old decision situation” and old “decision solutions”. For that reason, the approach considers two aspects to create learning environments in decision-making process: common terminologies or common language to express everything that concerns to decision-making problem and iterative procedure for structuring problem.

Definition of Common Terminologies: Ontology for Supply Chain

Decision-making process is a process that deals directly with uncertainty. Under decision-making scenario, decision maker needs to reduce the uncertainty in two different sources: the information about the situation under the analysis and the possible consequences of some actions as a result of decision-making process. To reduce uncertainty in both situations, decision maker needs lot information in almost of present approach. One problem creates another: decision maker needs information but the searching for that information generates the problem of processing a huge amount of information. At the first steps it is impossible to decide which information is relevant or not for decision-making, for that reason many organizations analyze and collecting a big quantity of information (from inside and outside organization). In many previous research [4,2,8,23,24,15] this fact is a common pattern in decision-making process. There are many examples of this: logistics (inventory management), marketing (customer loyalty), sales (distribution channel), finance (project feasibility), and others.

In supply chain environments partners have similar problem independently of the nature of their contributions in the chain. This conclusion is well supported due to that in the supply chain environments, partners need to coordinate logistics operation as a clock mechanism. Since the decisions in the supply chain are common, which is the reason that managers don’t share decision-making problem & solutions formulations? This problem has two different perspectives.

First of all there is not a common language to share the knowledge about how to handle decision-making problem. This means that managers don’t have a common language to share knowledge inside organization and even more, between other partners. Second, there is not a common procedure to formulate problem & solutions in decision-making that aid other managers in decision-making process.

Recently (1996) Supply Chain Council created SCOR model. This model created a standardization of terms and performance measurement in the supply chain operations. By sharing the SCOR terminology through supply chain it is possible to create a common language to express decision-making process. To create a common procedure that aids decision making process by sharing knowledge between supply chain partners is extremely necessary to create a common semantic base. SCOR acts as a common semantic base, but we need to express SCOR in some format that allows sharing through internetworking infrastructure of supply chain. Figure 2 shows a conceptual approach to express SCOR into RDF document. If SCOR model is expressed in RDF document, the semantic base have an easy way to understand for human and computer in decision-making analysis process.


Figure 2. SCOR in RDF documents. A conceptual design.

Using RDF document SCOR can be used as ontological base for supply chain decision-making process. Figure 3 shows a Protégé snapshot for SCOR ontology.

Figure 3. SCOR ontology. Iterative Procedure to Structuring Decision-Making Problem

Figure 3. SCOR ontology.
Iterative Procedure to Structuring Decision-Making Problem

Supply chain is a complex system. All complex systems need a more complex control system but few elements decide the behavior of whole system. This conclusion suggests that all complex systems (like supply chain) can be handled by few elements as well as the considerations of the special relationship among them in different situations.

Considering a decision-making as a system, the complexity of problem structuring would be a natural consequence of the previous definition. Organizations work in stable environments, with stable people and stable technology and that is the main thesis. Organizations face events (even internal and external) that force to change their stable situation. This situational context represents the antithesis and produces change in all organization. Under current conditions, the “old procedure”, solutions and thesis are not working any more. Thus, the organization moves and adapts to the new condition and this changes are the synthesis that will be the new thesis of the organization. This is a cycle over the time.

Based on this definition, we propose to use the following procedure for problem & solutions structuring. In figure 4 we show the cycle of decision-making process and with the use of the cause-effect modeling technique suggested by Goldratt (1999).


Decision-making problem structuring can be represented as cause-effect relationship between elements of the supply chain definitions. These elements come from SCOR ontology and in any particular situation this relationship can be characterized by relations between performances indicators (associate with process-elements in SCOR ontology) that represent non desired effect present in decision situation under analysis.

How this procedure could help managers in decision-making process? One of the most difficult task in decision-making process is to identify the underline structure of the problem. That means discovering how different factors (and which factors) have influence in a decision situation. This discovering procedure is like a regression procedure where analysts need to identify the variable, the relationship between variables and the relative importance of them in equation. In decision-making process, the performance indicators act as variables and the relationship between indicators characterize the behavior of the system. Figure 5 describe this procedure.


Due to indicators come from ontology and all indicators have a relationship with process-elements, examining the indicators relationship managers understand the relationship between process-elements or vice versa. When a situation occurs managers are able to identify which process-elements are interacting and based on this information managers try to identify possible solutions. However, indicators are the “red light” in management’s context. When some indicator is out of control, managers infer that something in the system are wrong and need a decision or need to take some corrective actions to stabilize the system. Additional mechanism could be incorporated to validate knowledge like Modified Petri Net (see [25,26]).

From one situation to another, managers learned by means of the analysis of the cause-effect relationship in each situation. After some time based on the characteristics of the situation (cause-effect relationship structure) managers infer what decision should be taken based on the effectiveness of previous decision. Sharing these relationships through all network point in the supply chain managers from other organization that belongs to the same chain learn from the other managers’ experience. The cause-effect relationship could be expressed in RDF document by using N3 formats.

This is the nature of information to support decision-making process. The cause-effect relationships structuring represent the common pattern of decision problem. By using RDF storage management system or RDBS[1] it is possible to creates a common repository of these patterns to share through the supply network partners. Now, the alignments of the decision in the supply chain management have an information system to support them, due to the common pattern of decision problem and also solutions can be shared. The strategic decision level has in the information system a tool to reach the strategic decision-making alignments.


The decision about the relevance and irrelevance of information for decision-making has a solution in this approach. Cause-effect relationships are the relevant information in decision-making context and by means of the analysis of these relationships, managers could learn. Sharing information now means sharing knowledge within organization and with other organization in the supply chain. Strategic alignments have knowledge a support and this is better than the information support. Besides this helps to reduce the problem associated with new interpretations for other managers in different decision levels inside and through organization in the supply chain.

Based on the cause-effect relationship it is possible to create a computational procedure to aids decision making process. This computational procedure could be combined with artificial intelligence to derive a methodology to create the inference mechanism to describe the structure of problem and solution in decision-making process.


We wanted to recognize the valuable contributions made by the following colleagues: Nico Brehm who contribute with many advice and valuable suggestions; Adrian González Oliva by its contribution to the theoretical foundations; Mario Bishen and Tania Machado who reviewed all the referring details to the language.



  1. McCormack, K. and K. Kasper, The extended supply chain: The statistical study. Benchmarking: An International Journal, 2002. Volume 9(Number 2). pp. 133.
  2. Biswas, S. and Y. Narahari (2003) Object Oriented modeling and decision support for Supply Chains. Electronic Article.
  3. Narahari, Y. and S. Biswas (2001) Supply Chain Management: Modelling and decision making. Electronic Article. A framework for Supply Chain Management and decision making models
  4. Arnott, D.R. (1998) A Framework for Understanding Decision Support Systems Evolution. Electronic Article.
  5. Geerts, G.L., W.E. McCarthy, and A. Andersen (2002) Modeling Business Enterprises as Value-Added Process Hierarchies with Resource-Event-Agent Object Templates. Electronic Article.
  6. Capurro, R., The concept of information, in Draft. 2002: págs. 1.
  7. Mosquera Saez, I. and E. Wong, JIT, TQM o TOC ¿Cómo afrontar el nuevo desafío? 1996, Universidad Central “Marta Abreu” de Las Villas. pp.1.
  8. Corner, J., J. Buchanan, and M. Heing, Dynamic decision problem structuring. 2001, University of Waikato, Tel Aviv University. pp. 1.
  9. Summers, J.D., B.M. McLaren, and D.W. Aha. Towards Applying Case-Based Reasoning to Composable Behavior Modeling. In Conference on Behavior Representation (BRIMS). 2004. Virginia.
  10. F.Leymann and D.Roller, Using flows in information integration. IBM System Journal, 2002. Vol.41(No.4). pp.732.
  11. George S. Nezlek, H.K.J.a.D.L.N., An integrated approach to enterprise computing architectures, in Communication of the ACM. 1999. pp. 82.
  12. Haberman, A.W.S.a.F., Making ERP a success, in Communication of the ACM. 2000. pp. 57.
  13. Keith, B.A., et al., Conceptual models for coordinating the design of use work with the design of information systems. Elsevier. Data&Knowledge Engineering, 2000. pags 191.
  14. King, J.L. and K. Lyytinen, Information system: the state of the field, ed. R. Boland and R. Hirschheim. 2006: John Wiley & Sons, Ltd. pags. 355.
  15. Tsoukiàs, A., On the concept of decision aiding process. 2003, Université Paris Dauphine: Paris. pp. 1.
  16. Becerra-Fernández, I., K.E. Murphy, and S.J. Simon, Integrating ERP in the businees school curriculum, in Communication of the ACM. 2000. pp. 39.
  17. Dissel, M.K.a.H.v., ERP system Migrations, in Communication of the ACM. 2000. pp. 52.
  18. Everdingen, Y.v., J.v. Hillegerberg, and E. Waarts, ERP adoption by Europe midsize companies, in Communication of the ACM. 2000. pp. 27.
  19. Markus, M.L., C. Tanis, and P.C.v. Fenema, Multisite ERP implementations, in Communication of the ACM. 2000. pp. 42.
  20. Sinhg, N., Unifying heterogeneous information models, in Communication of the ACM. 1998. pp. 37.
  21. Perales, J.C., A. Catena, and A. Maldonado. Aprendizaje de relaciones de contingencia y causalidad: Hacia un análisis integral del aprendizaje causal desde una perspectiva computacional. 2001. Departamento de Psicologia Experimental. : Universidad de Granada. pp:1-62.
  22. Palmero, M.L.R. La teoría del aprendizaje significativo. In Conference on Concept Mapping. 2004. Pamplona, Spain.
  23. Mosquera Saez, I., G. Hernández Pérez, and J. Marx Gómez, Decision Making Support in Supply Chains through Modified Petri Nets and Case-based Reasoning, in 8th Annual Global Information Technology Management Association World Conference. 2006: Napoles, Italy.
  24. Perner, P., T.B. Belikova, and N.I. Yashunskaya. Knowledge Acquisition by Decision Tree Induction for Interpretation of Digital Images in Radiology. In Advances in Structural and Syntactical Pattern Recognition. 1996. Berlin: Springer Verlag Lncs.
  25. Billington, J., et al. (2005) The Petri Net Markup Language: Concepts, Technology, and Tools. Electronic Article. pp.1
  26. Koriem, S.M., A Fuzzy Petri Net Tools for modelling and verification of knowledge – based systems. The Computer Journal, 2000. Vol. 43. No. 3. 2000. pp. 206

[1] See other reference for this.