Numerical Assignment Techniques Analysis

Numerical Assignment Technique (NAT):
The numeral assignment technique is based on the principle that each requirement is assigned a symbol representing the requirement’s perceived importance. This approach is common in Quality Function Deployment (QFD) where prioritizing of candidate requirements is required. Several variants based on the numeral assignment technique exist. A straightforward approach to the technique is presented by (Brackett, 1990), who suggest that requirements should be classified as mandatory, desirable, or inessential. An approach using finer granularity is to assign each requirement a number on a scale ranging from 1 to 5, where the numbers indicate:
5. Mandatory (the customer cannot do without it).
4. Very important

Azar, 2007). It provides visibility for all stakeholders during decision making, eliminating lengthy discussions and arguments over individual requirements by emphasizing the core business values. The first step in setting up a value oriented prioritization process is to establish a framework for identifying the business’s core values and the relative relationships among those values. VOP uses the relationships that exist between core business values to assess and prioritize requirements and ensure their traceability. The VOP framework establishes a mechanism for quantifying and ordering requirements for an application increment, a prototype, or a software requirements specification. Company executives identify the core business values and use a simple ordinal scale to weight them according to their importance to the

Methods based on AHP also allow the possibility of checking the consistency of the priorities. The conclusion from the inherent properties is thus that AHP is most powerful.

The objective measures in Table 2 show that AHP, priority group and bubble sort require the highest number of decisions (around 80) and spanning tree, VOP, NAT, CV requires the fewest (around 10). The number of decisions required for binary search and priority groups depend on the decisions taken during method execution, hence the three different values for each evaluator. Binary search, priority groups required around 30 decisions.

The total time consumption and the time consumption per decision are presented on an ordinal scale, as here we are only interested in the ranking of the methods. The results from the evaluation showed that AHP, NAT and binary search need the longest time to execute, while VOP, bubble sort and spanning tree were the fastest methods. If we divide the total time by the number of decisions, we see that binary search and spanning tree required most time per decision, while AHP and bubble sort were, on average, fastest per