A visual interactive goal programming-based decision support system: An empirical assessment

Visual interactive modelling (VIM) can contribute significantly to the building of models based on visualization of a problem from the user's perspective. In this study the VIM approach was used to develop a flexible and powerful user interface for a goal programming-based decision support system (DSS). Alter (1980) emphasized that, without a responsive and powerful interface, multiple criteria decision making (MCDM) methods such as goal programming (GP) can hardly be utilized by decision makers in spite of their conceptual relevancy and functional efficiency.;The study had two main objectives. The first was to develop a prototype GP-based DSS, using the VIM approach. This DSS aimed to assist decision makers in arriving at preferred solutions to multiple criteria problems with conflicting objectives. The proposed system was not restricted to a model or algorithm, but encompassed support strategies for comparison, preference elicitation, and choice. The second objective was to test the effect of this DSS compared to a traditional or non-visual DSS that addressed the same decision-making problem. Since assessing task-related factors through field study would be very difficult, if not impossible, a laboratory experiment was conducted to assess the impact of the proposed system and allow manipulation of task factors. A total of sixty-four second-year MBA students at Richard Ivey School of Business voluntarily took part in the experiment.;The results were evaluated on several dimensions, including performance, attitude, user's learning mode, user's interaction, efficiency of the model, decision process, individual differences in cognitive styles, and consistency in the final solution. In aggregate terms the VI-DSS was superior because it outperformed the non-VI DSS along almost all evaluation dimensions. Therefore, it was concluded that the VIM approach improved the effectiveness and efficiency of the system.;The contribution of this study was highlighted by developing the concept of VI-GP and demonstrating the value of integrating a GP model with a visual interactive problem-oriented interface. In other words, this study was an attempt to develop a GP-based VI model that visualized a managerial decision-making problem, rather than the algorithm of the problem, and use it to solve the problem. This study also provided a further empirical assessment of the VIM approach by using new technology in building a VI model. Finally, it provided management science/operations research (OR/MS) practitioners with new insight into the development of GP-based DSS and a new way of incorporating subjective criteria into optimization problems that facilitated the involvement of decision-makers' values and judgment in the problem-solving loop.