August 1998 Issues in Education: Modeling Software & the MBA OR/MS Course By By Matthew J. Liberatore and Robert L. Nydick There has been a rebirth of interest in the teaching of the first OR/MS course, driven in part by the reduction of required management science courses in MBA programs. [Report of the INFORMS Task Force, 1997.] Recently there has been a movement to use spreadsheets as the primary modeling vehicle. We describe a different approach that uses modeling software to empower students to create and implement their own models through group projects. A more in-depth discussion can be found in Liberatore and Nydick [1998] available on our Web site at www.vill.edu/~rnydick/home.html. We have successfully applied our approach to the required management science course in Villanova University's part-time MBA program. Our course has evolved over the past five years and includes modules in mathematical programming using LINGO and its modeling language, simulation using Extend and Stat::Fit, and decision analysis using Expert Choice. Why did we choose modeling software over spreadsheets? A variety of compelling reasons can be offered to support the use of spreadsheets including a large installed base of users, the use of spreadsheets as the primary analysis tool in many business situations, and the availability of add-ins for optimization and simulation [Savage, 1997]. Savage [1997] also discusses some of the disadvantages of spreadsheets, including documentability, scalability and hyperscalability. Documentability refers to the difficulty of documenting a spreadsheet model for ongoing maintenance and use. Scalability relates to the ability to change the number of items in a set, such as the number of products within a planning model. Hyperscalability considers the ability to add or remove dimensions, such as adding a time dimension to a static planning model. We agree that the use of spreadsheets is a step in the right direction. However, modeling software can overcome the disadvantages of spreadsheets. For example, we have used simulation add-ins and have found that it was difficult for the students to address queueing and process redesign situations that often occur in practice. Specifically, in modeling a supermarket, the discussion inevitably turns to such behaviors as customers balking and reneging and the changing patterns of arrivals and number of servers during the day. Currently available spreadsheet add-ins are unable to address these issues. Perhaps additional add-ins could be developed. However, we are interested in empowering our students with the capability to model these and other behaviors that are of interest to them. Visual simulation software packages, such as Extend, deliver this capability. An important and relevant issue is whether MBA students possess the requisite skills to take advantage of modeling software. Recently, software developers have become more attentive to ease of use and learning curve issues. Although spreadsheet modeling is easy to use initially, it quickly becomes more difficult for students to create more complex models. In contrast, once the student achieves a certain level of understanding with the modeling software, he or she can develop more comprehensive models. Our pedagogical approach enables students to learn the essentials of modeling with the software so that they can create their own models. We employ a building block approach that is similar to Savage's idea [1997, p. 44] of teaching students "how to assemble applications from decision objects." We use materials that help the student understand the operation of the software blocks and their relationship to modeling activities. After the first in-depth example in each module, the student quickly moves along the learning curve. He or she is then able to more easily model realistic situations, first in homework exercises, then in the midterm and final examinations, and finally in the two projects (a mathematical programming project and a decision analysis and/or simulation project). We emphasize how the various examples contain blocks that can be combined and/or modified to address other problems. Students easily identify project topics since: 1) many problems their companies face are simply variations of these models, and 2) many of the modeling blocks can be extended to other applications. Projects The completed projects are an extremely important outcome of this course. Students either conduct a pilot study using real company data, or completely model the actual situation. We consider a successful implementation to occur if either the project genuinely causes a positive change in the business, or if it demonstrates the relevance of using OR/MS through positive feedback from management. We have observed that projects tend to fall into the following categories: implemented immediately (about 25 percent); a very good chance of being implemented in the near future (about 25 percent); a chance of being implemented (about 25 percent); and very little chance of being implemented (about 25 percent). We now request that all student teams include a letter from management commenting on the value of completed projects. One can endlessly discuss and theorize about the advantages of any pedagogical approach. In our opinion, the "proof of the pudding" is in the application. We believe that our approach empowers part-time MBA students to produce, and often implement, quality management science projects. References 1. Liberatore, Matthew J. and Nydick, Robert L., 1998, "Breaking the Mold: A New Approach for Teaching the First MBA Management Science Course." Under second review at Interfaces. 2. Report on the Operating Subcommittee of the INFORMS Business School Education Task Force, 1997, "OR/MS and MBAs," OR/MS Today, Vol. 24, No. 1, pp. 36-41. 3. Savage, Sam, 1997, "Weighing the PROS and CONS of Decision Technology in Spreadsheets," OR/MS Today, Vol. 24, No. 1, pp. 42-48. Table of Contents OR/MS Today Home Page OR/MS Today copyright © 1998 by the Institute for Operations Research and the Management Sciences. All rights reserved. Lionheart Publishing, Inc. 506 Roswell Street, Suite 220, Marietta, GA 30060, USA Phone: 770-431-0867 | Fax: 770-432-6969 E-mail: lpi@lionhrtpub.com URL: http://www.lionhrtpub.com Web Site © Copyright 1998 by Lionheart Publishing, Inc. All rights reserved.