Let Smart Ants Help You Reduce the Delay Penalty of Multiple Software Projects

摘要

Delays often occur in real-world software development projects and may cause significant monetary penalties to software companies. Meanwhile, industry lessons have shown that adding inexperienced employees would cause further delays due to the learning curve and communication overhead. However, if employees with same or similar skills and domain knowledge can be rescheduled from other concurrent projects to help with the delayed projects, it may be possible to reduce or even eliminate delay penalties without requesting extra employees. Here, the big challenge is how to conduct employee rescheduling without having employees working overtime, which is an NP hard problem in nature. To address such a problem, this paper proposes a novel employee rescheduling strategy based on improved ant colony optimization algorithm. Specifically, three generic rules are proposed to improve the effectiveness in generating valid solutions. Preliminary results on benchmark projects show that our strategy can achieve much better effectiveness than its genetic algorithm based counterpart in reducing the overall delay penalty of multiple software projects.