Hyper-heuristics using multi-armed bandit models for multi-objective optimization

Almeida, Carolina; Gonçalves, Richard A.; Venske, Sandra M.; Lüders, Ricardo; Delgado, Myriam Regattieri

doi:10.1016/j.asoc.2020.106520

Cited by 22 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bonab et al (2019) designed a swarm-based simulated annealing hyper heuristic algorithm to deal with clustering problem. Almeida et al (2020) proposed a MOEA/D-based hyper heuristic algorithm and demonstrated that the algorithm can be well applied in a permutation flow shop scheduling problem. Chen et al (2021) proposed a hyper heuristic algorithm-based ensemble genetic programming method for solving stochastic resource constrained project scheduling problem by evolving an ensemble of priority rules.…”

Section: Optimization Methodsmentioning

confidence: 99%

A shuffled frog leaping-based hyper heuristic algorithm for material feeding scheduling optimization in an automatic monorail shuttle system

Zhou

Zhao

2023

Preprint

View full text Add to dashboard Cite

In recent years, mixed-model assembly lines (MMALs) have been widely adopted by the automotive enterprises to face the requirements of mass customization. Meanwhile, the diversified components of various end products are placing the material feeding processes at great challenges. Under this circumstance, to raise the material feeding efficiency, this paper proposes an automatic monorail shuttle system (AMSS) based on line-integrated supermarkets applied to the MMALs by introducing the load-exchangeable shuttles and crossovers. Considering the importance of cost-control in manufacturing enterprises, we establish a mathematical model with the objective to minimize the total costs of the material feeding system, which includes the installation costs of crossovers and the input and operation costs of load-exchangeable shuttles. Then, due to the NP-hard nature of the proposed problem, a shuffled frog leaping-based hyper heuristic (SBHH) algorithm is developed to determine the allocation and scheduling of the shuttles and crossovers, which adopts the shuffled frog leaping algorithm (SFLA) as the high-level heuristic to select the low-level heuristics. To improve the performance of the algorithm, the concept of ‘dynamic decision unit’ is presented to raise the solution accuracy and the convergence speed. Finally, the simulation results verified the superiority of the proposed SBHH algorithm both in solution quality and convergence speed by comparing with other optimization algorithm in the existing literature.

show abstract

Section: Optimization Methodsmentioning

confidence: 99%

A shuffled frog leaping-based hyper heuristic algorithm for material feeding scheduling optimization in an automatic monorail shuttle system

Zhou

Zhao

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A typical example of studying decision-making in reinforcement learning is the multi-armed bandit problem (MAB). It is originally a stateless mathematical model extracted from the scene of a multi-arm slot machine in a casino [22]. Pulling one arm of the slot machine may generate positive revenue.…”

Section: Reinforcement Learningmentioning

confidence: 99%

Dynamic pricing of differentiated products with incomplete information based on reinforcement learning

Wang

Cui

et al. 2022

IET Collab Intel Manufact

View full text Add to dashboard Cite

With the rapid development of the social economy, consumer demand is evolving towards diversification. To satisfy market demand, enterprises tend to improve competitiveness by providing differentiated products. How to price differentiated products becomes a hot topic. Traditionally, customers' preferences are assumed to be independent and identically distributed. With a known distribution, companies can easily make pricing decisions for differentiated products. However, such an assumption may be invalid in practice, especially for rapidly updating products. In this paper, a dynamic pricing policy for differentiated products with incomplete information is developed. An adaptive multi‐armed bandit algorithm based on reinforcement learning is proposed to balance exploration and exploitation. Numerical examples show that the frequency of price adjustment affects the total profit significantly. Specifically, the more chances to adjust the price, the higher the total profit. Furthermore, experiments show that the dynamic pricing policy proposed in this paper outperforms other algorithms, such as Softmax and UCB1.

show abstract

“…A memory mechanism that contains multiple initial solutions is introduced to retain the diversity of the solution. The MAB and relay hybridization technology are introduced to single-point search [41,42] . During the single-point search, the switching of two searching methods is determined by the probability p r , which is calculated as follows:…”

Section: Proposed High-level Strategymentioning

confidence: 99%

A Novel Cooperative Multi-Stage Hyper-Heuristic for Combination Optimization Problems

Zhao

Cao

et al. 2021

Complex Syst. Model. Simul.

View full text Add to dashboard Cite

A hyper-heuristic algorithm is a general solution framework that adaptively selects the optimizer to address complex problems. A classical hyper-heuristic framework consists of two levels, including the high-level heuristic and a set of low-level heuristics. The low-level heuristics to be used in the optimization process are chosen by the high-level tactics in the hyper-heuristic. In this study, a Cooperative Multi-Stage Hyper-Heuristic (CMS-HH) algorithm is proposed to address certain combinatorial optimization problems. In the CMS-HH, a genetic algorithm is introduced to perturb the initial solution to increase the diversity of the solution. In the search phase, an online learning mechanism based on the multi-armed bandits and relay hybridization technology are proposed to improve the quality of the solution. In addition, a multi-point search is introduced to cooperatively search with a single-point search when the state of the solution does not change in continuous time. The performance of the CMS-HH algorithm is assessed in six specific combinatorial optimization problems, including Boolean satisfiability problems, one-dimensional packing problems, permutation flow-shop scheduling problems, personnel scheduling problems, traveling salesman problems, and vehicle routing problems. The experimental results demonstrate the efficiency and significance of the proposed CMS-HH algorithm.

show abstract

Hyper-heuristics using multi-armed bandit models for multi-objective optimization

Cited by 22 publications

References 46 publications

A shuffled frog leaping-based hyper heuristic algorithm for material feeding scheduling optimization in an automatic monorail shuttle system

A shuffled frog leaping-based hyper heuristic algorithm for material feeding scheduling optimization in an automatic monorail shuttle system

Dynamic pricing of differentiated products with incomplete information based on reinforcement learning

A Novel Cooperative Multi-Stage Hyper-Heuristic for Combination Optimization Problems

Contact Info

Product

Resources

About