Читать книгу Digital Transformation: Evaluating Emerging Technologies - Группа авторов - Страница 35

Hierarchical Decision Modeling (HDM) is a technical tool used in project selection, resource allocation and evaluation decisionmaking. Its objective is to assist the user, by a series of pairwise comparisons, to reach quantifiable judgmental value using ratio scales. The underlying assumption is that each decision has a number of perspectives and each perspective has number of criteria to consider [1].

Thus, combining the perspectives—quantifiable or non-quantifiable—and the supporting criteria will help in determining the strategy (decision). The HDM is a process using multi-level decisions and utilizing multiple criteria by separating the overall system into several hierarchical levels.

The HDM is also a process based on reaching out to an independent panel of selected experts, who responds to questions by dividing 100 points between two alternatives at a time. The allocation of the points represents each expert’s independent judgment with respect to a specific criterion. The 100-point scale is from 1 to 99. The zero value is avoided to eliminate mathematical difficulties; however, if such a consideration is given, each expert selects 50 points. This means the judgment is neither important nor unimportant [1].

The HDM is based on a pairwise comparison analysis using linear algebra and matrix analysis. The goal is to find the eigenvalue and the eigenvector for each consideration in the matrix. In other words, pairwise comparison is a method used to determine how to evaluate alternatives by providing an easy and reliable means to rate and rank decision-making criteria. Weights are used and assigned to criteria and the results are normalized. The comparison is implemented in two stages:

1.Determining qualitatively which criteria is more important (i.e., establish a rank order of the criteria), and

2.Assigning a quantitative weight to each criterion, such that the qualitative rank order is satisfied.

The process is based on three steps that differ in their underlying scale. First, the measurement is based on a range from an ordinal perspective (i.e., weighting by ranking). The second step is about constructing an interval by weighted ranking, while the third step calculates the ratio scale—the pairwise comparison value. The three steps are summarized below, based on the document HDM by Dundar Kocaoglu.

•Step 1—Completion of the pairwise comparison matrix: Two considerations are evaluated at a time in terms of their relative importance. Index values from 1 to 99 are used. If criterion A is exactly as important as criterion B, this pair receives an index of 1. If A is much more important than B, the index is 99. All degrees are possible in between when comparing A to B. For a “less important” relationship, the fractions would be closer to 50 points. The values are entered row by row into a cross-matrix. The diagonal of the matrix contains only values of 1. The right upper half of the matrix is filled until each criterion has been compared to every other one [1].

•Step 2—Calculating the criteria weights: The weights of the individual criteria are calculated. First, a normalized comparison matrix is created: each value in the matrix is divided by the sum of its column. To get the weights of the individual criteria, the mean of each row of this second matrix is determined. These weights are already normalized; their sum is 1.

•Step 3—Assessment of the consistency matrix: A statistically reliable estimate of the consistency of the resulting weights is made.