Assignment of trainees to software project requirements: A stable matching based approach

• Trainee allocation to software projects is important from cost and satisfaction viewpoints.
• A stable matching based approach is proposed to improve the allocation performance.
• The proposed approach increases trainee satisfaction compared to prior approaches.
• We show how to reduce the unstable assignments and its positive implications.

Greedy and optimization approaches used for assigning trainees to software project requirements can give rise to unstable pairs. The presence of unstable pairs can unnecessarily hamper both trainee and project requirement satisfaction. In this paper, we use the theory of stable matching to remove the unstable pairs. We propose an efficient method to predict preferences for both trainees and project requirements using the utility theory. Our computational experiments suggest that the stable matching approach performs better with average trainee and project requirement satisfaction in terms of the preference ranks of allocated choices. Compared to cost based optimization model, the proposed approach obtains significantly better trainee project match at the expense of small additional allocation cost.