Enterprise service composition in IIoT manufacturing: integer linear optimization based on the hybrid multi-objective grey wolf optimizer

Nowadays, optimal value creation by smart awareness in the composition of enterprise business services in an architecture-centric way in industrial internet of things (IIoT) manufacturing is a leading trend. In this research, as the first innovation, integer linear optimization modeling of enterprise service composition (ESC) as an NP-hard problem is performed while observing the structure of service composition in enterprise architecture. The proposed model optimizes ESC based on the hybrid multi-objective grey wolf optimizer (HGO), which is also considered an innovation in the mathematical model by using evaluator operators (crossover and mutation). Inverted generational distance (IGD), Spacing, Spread, and also average execution time are discussed in this paper as four main evaluation factors. HGO is compared with three famous multi-objective optimization (MOO) algorithms (NSGA II, MOPSO, and MOEAD). Finally, the proposed model’s Pareto front (PF) shows a better convergence than the other ones. The developed ESC model was applied to 15 standard optimization test problems compared with the three optimization algorithms. The results showed that HGO improved the optimality rate of enterprise services in IIoT manufacturing. The proposed model was executed from 100 to 1000 times in the production facility petrochemical industry as a case study, and the outcome of the diagram showed between 0.2 and 0.4 that which means a more desirable optimal rate result than the three famous multi-objective optimization algorithms.