The 0/1 Multidimensional Knapsack Problem and Its Variants: A Survey of Practical Models and Heuristic Approaches

Abstract

The 0/1 Multidimensional Knapsack Problem (0/1 MKP) is an interesting NP-hard combinatorial optimization problem that can model a number of challenging applications in logistics, finance, telecommunications and other fields. In the 0/1 MKP, a set of items is given, each with a size and value, which has to be placed into a knapsack that has a certain number of dimensions having each a limited capacity. The goal is to find a subset of items leading to the maximum total profit while respecting the capacity constraints. Even though the 0/1 MKP is well studied in the literature, we can just find a little number of recent review papers on this problem. Furthermore, the existing reviews focus particularly on some specific issues. This paper aims to give a general and comprehensive survey of the considered problem so that it can be useful for both researchers and practitioners. Indeed, we first describe the 0/1 MKP and its relevant variants. Then, we present the detailed models of some important real-world applications of this problem. Moreover, an important collection of recently published heuristics and metaheuristics is categorized and briefly reviewed. These approaches are then quantitatively compared through some indicative statistics. Finally, some synthetic remarks and research directions are highlighted in the conclusion.

Intelligent system for recruitment decision making using an alternative parallel-sequential genetic algorithm

“…To clarify this point, in the case of small recruitment instances, we use the modified version of the standard genetic algorithm (SGA) to obtain the improved sequential version (SeqGA), the modifications, are made at the level of generation of the initial population, the crossover and 'insertion operators. The results obtained using SeqGA are satisfactory which allows to right decision recruitment [15,16]. In the case of large instances or when adding the constraints to recruitment problem as cost post constraints, we can use a parallel genetic algorithm PGA through a multiprocessor architectures.…”

Heating Homes with Servers: Workload Scheduling for Heat Reuse in Distributed Data Centers

“…The second one determines the workload scheduling matrix such that a large enough workload is scheduled for execution on each server to generate heat to meet the demand. The latter problem can be solved using an adaptation of the Multiple Knapsack Problem [ 72 ].…”

Bitrate Adaptation for Video Streaming Services in Edge Caching Systems

“…Constraints (11) and (12) are the equivalent forms of (2) and (3). The structure of (P 2 ) strictly follows the typical structure of MMKP [24] and hence, it is NP-hard. We propose the OIGA algorithm to solve (P 1 ) with a minimal optimality gap compared to branch-and-bound (BB) method.…”