Outsourced Products Task Allocation via Group Multirole Assignment with Considering Variance

The phenomenon of uneven product quality in software outsourcing enterprises is common. Even for products with similar functionality, the company cannot guarantee the final delivery quality. Effectively solve the stability of product quality, which is more conducive to long-term cooperation between customers and the company. Hence, this paper formalizes the Outsourced Products Task Allocation (OPTA) problem via the Environments-Classes, Agents, Roles, Groups, Objects (E-CARGO) model. Through its sub model Group Multirole Assignment (GMRA), a team with the optimal qualification value can be obtained. However, the stability of the assignment result is still not guaranteed. Therefore, this paper innovatively introduces the concept of variance and proposes the Group Multiple Role Assignment with Considering Variance (GMRACV) model to tackle this issue. And make better improvements to it, achieving a performance loss of 1% in exchange for about 32% stability. Large-scale randomized experiments show that the proposed model can effectively reduce the variance between products while maintaining the overall quality of all products. And for data with different distributions, the model can still obtain excellent results stably, which further verifies the feasibility of the model.