Date of Award

1992

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Abstract

Visual interactive simulation (VIS) was first developed in the mid-1970s and has been claimed to be a better type of simulation model than traditional simulation for supporting decision making. VIS models which compare two simulated systems using paired-difference statistics have been claimed to be a more powerful decision-support tool than viewing animated simulation model output or steady state statistics. Since simulation is one of the most frequently used techniques in decision support systems (DSS), an examination of which type of simulation model is better for developing DSS is of vital importance. This dissertation focuses on examining and comparing the relative effectiveness and efficiency of three types of simulation models: traditional simulation models, conventional VIS models, and VIS models with paired-systems and paired-difference statistics.;The research was done through a laboratory experiment in which seventy-one second-year Masters' students in Business Administration at the Western Business School participated and were randomly assigned to one of the three experimental groups. Three DSS (one based on a traditional simulation model, one on a conventional VIS model, and one on a VIS model with paired-systems and paired-difference statistics) were developed and each was assigned to one of the three experimental groups. Subjects were asked to solve a production problem presented as a case using the DSS provided.;The results of the experiment indicated that of the three types of DSS, the one based on a VIS model with paired-systems and paired-difference statistics was the most effective and efficient. The DSS based on a conventional VIS model was the second most effective, while the traditional simulation was the least effective. In particular, the DSS based on a VIS model with paired-systems and paired-difference statistics significantly outperformed traditional simulation on all five evaluation criteria defined and used in this study.;This research made three important contributions: first, the work provided management science/operations research (MS/OR) researchers with the first empirical comparison of three different types of simulation-based DSS, and the results of the study provided the first strong empirical case for the use of VIS. Second, the research rigorously tested two claims in the literature by proponents of VIS. Third, the results of this research provided MS/OR practitioners with new insight on the development of simulation-based DSS.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.