Advancing Economic Management through Multi-Objective Optimization Algorithms: Design and Implementation of an Innovative Information System

Abstract

This paper addresses the critical risks associated with ERP implementation—including organizational, professional skill, project management, system, user, and technical risks—by proposing a novel ERP management information system design utilizing a multi-objective optimization algorithm. We detail the interconnections between service modules, components, process modules, and activity modules, utilizing mathematical modeling to articulate four decision-making objectives integral to ERP implementation. These objectives are subsequently optimized using the Multi-Objective Particle Swarm Optimization (MOPSO) algorithm. Our empirical analysis, conducted over 1000 iterations of the MOPSO algorithm, demonstrates that most solution particles converge within a region characterized by profit P ranging between $2,850,000 and $2,950,000, a standard deviation ? of [0.35, 0.65], and an average working efficiency ? of [70%, 85%], signifying robust stability and the formation of a Pareto front for the model. The optimal compromise solution achieved notable target values: profit P = $2,966,140, standard deviation ? = 0.31, overdue hours h = 37, and average working efficiency ? = 89.39%. The findings validate the efficacy of applying a modularization approach in designing knowledge-based services, and a multi-objective optimization framework for quantitative control. This strategy not only defines service scopes with precision—using the service process module as a boundary—but also dynamically adjusts service costs based on customer demand. This enables efficient resource allocation within the service supply chain, enhancing overall service efficiency. Such methodologies are crucial in biotechnological environments where data-driven decision-making is key to operational success.