Issues in the design, development and use of process management models

Shantanu Bose, Purdue University


This thesis proposes that detailed models of business processes must be used to make decisions in a supply chain. Traditionally, supply chains have been studied by three communities of researchers, those from Management Science, from Industrial Engineering and from Chemical Engineering. While, the first two groups fail to use detailed plant-level descriptions in the supply chain studies, the last group ignores the economic objectives. This dissertation show how to marry the two types of models, viz., higher level economic models and lower level detailed models. To that end, two modeling platforms, ePIMA: Environment for Process Investigation and Management Analysis, and, GNP: Generalized Model Predictive Framework, have been introduced as modeling approaches. A three-tiered supply chain, consisting of a supplier, a manufacturer and several retailers, is studied. This problem was motivated by the fast moving consumer goods industry. Part of the data used in this thesis has been obtained from an industrial sponsor, but it has been sanitized before usage. The rest of the data has been generated since the original industrial problem was enhanced to include further research issues. In this thesis, the author assumes the role of the manufacturer, and determines what decisions and actions on the part of the manufacturer will ensure the best performance in the supply chain. The performance is measured as profit realization and demand fulfillment. Two categories of issues have been studied. The first category involves the coordination between retailers and the manufacturer, and manufacturer and supplier. The second category involves design-type decisions within the manufacturing facility. In both these categories, uncertainty in demand has been included and studies have been done for “good”, “average” and “bad” forecast qualities. In addition to the uncertainty due to non-ideal forecasts, the effect of uncertainties in the production process, such as batch-processing time and batch-process yield, have been studied and solutions to combat them are presented. Using the two computational frameworks, this dissertation demonstrates how the models can be used in a supply chain to make decisions. This dissertation also provides a comprehensive description of many relevant issues in the area of supply chain management. A subset of these issues have been used as examples to show what results and conclusions can be obtained by using detailed models of business processes.




Reklaitis, Purdue University.

Subject Area

Chemical engineering|Industrial engineering|Operations research|Management

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server