Multi-objective newsboy problem with random-fuzzy demand

Abstract

In recent years, the entire nation has expressed deep concern over the escalating greenhouse gas emissions that contribute to global warming. In light of this, we have developed a model for a single-period multi-objective inventory problem, taking into account the uncertainties associated with perishable goods. The primary objectives of this formulation are to maximize expected profit while simultaneously minimizing the total carbon emissions. On the other hand, given the growing complexity and challenging nature of the newsboy problem in an uncertain environment, we have made an assumption that the product’s demand follows the distribution of random-fuzzy variables. To address this, we employ credibility theory to transform the random-fuzzy variable into a crisp equivalent form, enabling more effective analysis and decision-making. Due to the NP-hard nature of the developed problem model, the proposed approach for solving it involves the utilization of Multi-objective Particle Swarm Optimization (MOPSO), a multi-objective evolutionary algorithm. Furthermore, as there is a lack of existing benchmarks in the literature for verifying and validating the obtained results, we propose the utilization of a Non-dominated Sorting Genetic Algorithm-II to address the above-mentioned issue. Subsequently, we compare the performance of the two algorithms using various multi-objective performance measures. The comparison reveals that the MOPSO algorithm emerges as the better choice for solving our proposed model. In conclusion, a sensitivity analysis has been conducted by varying the uncertainty parameter, enabling the decision-maker to adjust their order based on their specific demand.