Chance-Constrained Multiple-Choice Knapsack Problem: Model, Algorithms, and Applications
Xuanfeng Li,Shengcai Liu,Jin Wang,Xiao Chen,Yew-Soon Ong,Ke Tang
DOI: https://doi.org/10.48550/arXiv.2306.14690
2023-12-15
Abstract:The multiple-choice knapsack problem (MCKP) is a classic NP-hard combinatorial optimization problem. Motivated by several significant real-world applications, this work investigates a novel variant of MCKP called chance-constrained multiple-choice knapsack problem (CCMCKP), where the item weights are random variables. In particular, we focus on the practical scenario of CCMCKP, where the probability distributions of random weights are unknown but only sample data is available. We first present the problem formulation of CCMCKP and then establish two benchmark sets. The first set contains synthetic instances and the second set is devised to simulate a real-world application scenario of a certain telecommunication company. To solve CCMCKP, we propose a data-driven adaptive local search (DDALS) algorithm. The main novelty of DDALS lies in its data-driven solution evaluation approach that can effectively handle unknown probability distributions of item weights. Moreover, in cases with unknown distributions, high intensity of chance constraints, limited amount of sample data and large-scale problems, it still exhibits good performance. Experimental results demonstrate the superiority of DDALS over other baselines on the two benchmarks. Additionally, ablation studies confirm the effectiveness of each component of the algorithm. Finally, DDALS can serve as the baseline for future research, and the benchmark sets are open-sourced to further promote research on this challenging problem.
Neural and Evolutionary Computing