Cits5502 | Software Processes | Assessment Answers

For the following tasks, select at least 3 metrics for measuring the test and repair process. Some possible metrics for showing progress in each week are:

• fix defects in random order
• fix easy defects first
• fix hard defects first
• fixmajordefects first
• fix minor defects first
• fix defects in the order they are found
• . . . something else (describe carefully)

As an example: you might include a graph showing what a single metric looks like, for a fixed staff allocation strategy, for each of four different defect prioritization strategies.

Include your tables or graphs in a report section in which you state your objective in performing these simulations, identify anyassumptionsyou needed to make, and present the results. Also state any limitations in your approach.

Tasks:

Now, from the above graph it is seen that two curves are fitted here for the actual data which have the R^2 values 84.98% and 92.81% respectively. The linear regression fitted has explained 84.98% variation in the actual data and exponential line fitted has explained 92.81% of variation in the actual data. Hence, the exponential fitted line is the best fit for this defects vs weeks datasets and the goodness of it coefficient is 92.81%.

3. Many strategies can be employed to find and fix the defects in each week with three persons that is with 75 person hours in a week. The best strategy which is considered to find and fix the defects is to employ 1 person for finding the defects and two persons fixing them. Now, the four strategies that are selected which follows the common strategy rule of employing one person for finding the defects and two persons fixing those are

1) Fixing the easy defects first by both of the two persons and then move to the hard defect fixing (major easy will be prioritized first).

2) Fixing the hard defects first and then fixing the easy defects by both persons (major hard defects will be prioritized first).

3) Fixing the major defects first and then moving to fix the minor defects (major easy will be prioritized first).

4) Fixing the minor defects first and then moving to fix the major defects (minor easy defects will be prioritized first).

Now, the three metrics which will be measured by employing each of the above four strategies to make the decision about the most efficient strategy for finding and fixing the defects in the system are given below.

1) The overall impact of found defects that are to be fixed where the major defect is found to have 7 times more than the impact of a minor one.

2) The number of unfixed defects including both found and not found.

3) The ratio of fixed defects and found defects.

4. Now, the three metrics Impact, unfixed defects and ratio of found vs fixed defects at the end of a week is calculated for each of the strategies. The comparison of the four strategies for the three metrics ratio, unfixed defects and impact are shown below.

Ratio of fixed vs found defects:

Number of unfixed defects:

Impact of defects that are still to be fixed:

Hence, by comparison of the four strategies it can be seen from the graphs that the strategy with the lowest impact is fixing the major defects first. The strategy with lowest number of unfixed defects is fixing the easy defects first and the strategy with highest number of fixed and total defects ratio is fixing the easy defects first. Hence, as the strategy of fixing the easy defects first has the most beneficial for two metrics out of three, hence the best strategy is to fix the easy defects first.

The limitation of this approach is that it is assumed that efficiency of the both testers who are fixing the defect are same and tester who finds the defects has the same defect detection rate as given in the data where total of 300 defects were detected and fixed in 20 weeks period. This assumption may not hold true in the actual case and hence the results and the graphs may vary according to that.

5. Now, from the calculation of workers’ allocation in defect finding and fixing it can be concluded that from the above graph the four strategies will produce the more or less identical results but the metrics may be different. The most efficient strategy based on customers reliability and satisfaction is to reduce average time for major defects, minimizing the impacts and maximizing the ratio (defects found/defects repaired). Hence, satisfying the above three metrics it can be found that the most efficient strategy is to allocate 2 persons for fixing and fixing the easy defects with major impacts first and then moving to the hard and major, finally ending with fixing the minor defects. More, specifically as found from the above graphs the most beneficial strategy is to fix the easy defects first. The performance of the 3 team member working in the above strategy is expected to be improved in the coming weeks as more number of major defects will be removed from the system.