Questionnaire design improvement and missing item scores estimation for rapid and efficient decision making

Abstract

In unconventional emergency decision making process using the analytic hierarchy process (AHP), it is important to quickly collect and process experts’ opinions to make a rapid decision. Questionnaire survey is a commonly used way to collect opinions and views in the AHP. However, many factors such as tedious design format, redundant content, and long length, may lead to inconsistent comparison matrix for the decision problem. Invalid or bad results of a questionnaire survey may cause the decision makers to make wrong decision. Furthermore, in the AHP, the score items for a comparison matrix in a questionnaire increase drastically if there are more comparisons, which result in longer survey. In this paper, a scale format is used to design the score items for a comparison matrix in questionnaire survey. Besides, an induced bias matrix model (IBMM) is proposed to estimate the missing item scores of the reciprocal pairwise comparison matrix. The survey questionnaire can be improved according to the importance of score items and emergency degree of the surveyed questions. A numerical example is used to illustrate the proposed method in unconventional emergency decision making. In addition, three cases of this example are analyzed and compared to address the effectiveness and feasibility of the proposed estimation model in the survey questionnaire design.