Resource allocation model in software development
B. Luong, Dar-Biau Liu · 2002
In this paper, we apply the queueing theory to model software management and maintenance during the system test, alpha test, and the beta test phases of a software system. The model helps us to verify the progress of the testing phase and estimate its cost. Based on the statistical results derived from the model, we then make a decision about changes in the workforce early rather than letting the product miss the scheduled deadline. Moreover, the model helps developers and managers to measure the product reliability, the quality of service, and to accurately estimate the number of developers needed for product maintenance. Three queue models, M/M/1, M/M/c, and the Jackson network are discussed in this paper. The basic unknown parameters of the models are the fault-report rate and the fixing-fault rate. We also present the technique to estimate the fault report and fixing fault intensities. The model gives us some information about workforce utilization, the average number of fault reports in the queue and system, the response time to customers, and software team missed deadlines. Finally we compare the results obtained from the statistical analysis of real data collected during the testing phase of a computer telephony integration system.