ERIC Number: EJ1061634
Record Type: Journal
Publication Date: 2013-Jul
Abstractor: As Provided
Reference Count: N/A
An Active Learning Exercise for Introducing Agent-Based Modeling
Pinder, Jonathan P.
Decision Sciences Journal of Innovative Education, v11 n3 p221-232 Jul 2013
Recent developments in agent-based modeling as a method of systems analysis and optimization indicate that students in business analytics need an introduction to the terminology, concepts, and framework of agent-based modeling. This article presents an active learning exercise for MBA students in business analytics that demonstrates agent-based modeling by solving a knapsack optimization problem. For the activity, students act as naïve agents by using dice to randomly selecting items for a finite capacity knapsack to maximize the value of the knapsack. Students then design a greedy heuristic to skew the probability of selection item. These pencil-and-paper models are then implemented in a spreadsheet model to demonstrate the effects of altering the agents' behavior. Finally, a binary integer programming model is examined to contrast agent-based modeling with traditional mathematical programming formulations. This exercise is innovative because it combines student engagement via active learning with an innovative, individual-based, modeling methodology.
Descriptors: Active Learning, Business Administration Education, Graduate Students, Masters Programs, Assignments, Models, Statistical Analysis, Heuristics, Probability, Spreadsheets, Programming, Learner Engagement
Wiley-Blackwell. 350 Main Street, Malden, MA 02148. Tel: 800-835-6770; Tel: 781-388-8598; Fax: 781-388-8232; e-mail: email@example.com; Web site: http://www.wiley.com/WileyCDA
Publication Type: Journal Articles; Reports - Descriptive
Education Level: Higher Education; Postsecondary Education
Authoring Institution: N/A