A new fuzzy QoS-aware manufacture service composition method using extended flower pollination algorithm

With the increasing complexity of manufacturing tasks and the exponential growth of candidate services, manufacturing service composition has become considerably challenging in relation to the integration of service supply chains in fuzzy manufacturing environments. Quality of service (QoS), as a popular index, is widely used to evaluate the fitness of solutions to the manufacturing service composition (SMSC). In this study, we first establish a new fuzzy QoS-aware mathematical model that considers the preferences of manufacturing enterprises by assigning different sub-tasks with different weights to evaluate the global fuzzy QoS of the SMSCs. We then extend the flower pollination algorithm (FPA) to obtain an optimal SMSC more effectively by making the switch probability self-adaptive, improving the local search ability, and adding the strategy of elite replacement. Finally, we demonstrate that the proposed extended FPA is an effective and efficient algorithm for solving the manufacturing service composition problem with differently weighted sub-tasks in a fuzzy manufacturing environment. We do this by comparing it with other well-known metaheuristic algorithms such as basic FPA, genetic algorithm, cuckoo search algorithm, and particle swarm optimization.