Dynamic hierarchical collaborative optimisation for process planning and scheduling using crowdsourcing strategies

Platform-based crowdsourcing manufacturing has recently garnered wide attention as it is a business model that facilitates economies of scale and cost efficiency in production. The inherent coupling of process planning and production scheduling (PPPS) in a platform-based crowdsourcing manufacturing environment necessitates collaborative optimisation of PPPS decisions. Existing research that assumes PPPS decisions are integrated into one static single-level optimisation problem becomes no longer applicable with the arrival of the crowdsourcing mode. This paper presents a dynamic hierarchical collaborative optimisation (DHCO) mechanism that considers a process planning to interact with scheduling according to the optimal decision of the open manufacturing platform. A bilevel mixed 0-1 nonlinear programming model is established with the platform acting as the leader and the manufacturing enterprises serving as the follower. It is solved by a nested genetic algorithm (NGA). A case study of a part family is presented to illustrate feasibility of DHCO. Through comparative experiments, it is found that integrating crowdsourcing strategies into process planning activities is advisable for a platform to increase competitive advantages. The proposed model can manage well the conflict and collaboration between PPPS and balances the benefits of a platform with the manufacturing enterprise impacts triggered by planning activities. Abbreviations: DHCO: Dynamic Hierarchical Collaborative Optimisation; IOM: Integrated Optimisation Method; KKT: Karush-Kuhn-Tucker; MNL: Multinomial Logit; NGA: Nested Genetic Algorithm; PFI: Process Flexibility Index; PPPS: Process Planning and Production Scheduling; PSI: Process Similarity Index; TOM: Two-stage Optimisation Method.