Hybrid Shuffled Frog Leaping Algorithm for Batch Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional batch scheduling methods fail to effectively consider job deterioration and often converge to local optima, leading to unreasonable scheduling in parallel-batching processing machines.
Innovation Solution
A hybrid method combining the shuffled frog leaping algorithm with variable neighborhood search, initializing frog populations, calculating fitness values, and employing neighborhood search strategies to improve scheduling rationality and convergence accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional shuffled frog leaping algorithm is used for batch scheduling, then the algorithm can perform local search and information exchange, but the convergence accuracy is low and it easily plunges into local optimum
Solution Approach 1:
The patent combines the shuffled frog leaping algorithm with variable neighborhood search algorithm to create a hybrid algorithm. The SFLA provides population-based search and information exchange, while VNS adds multiple neighborhood structures and local search strategies. This merging resolves the contradiction by maintaining the global search capability of SFLA while enhancing local search precision through VNS, thereby improving convergence accuracy without sacrificing scheduling rationality.
Solution Approach 2:
The hybrid algorithm acts as a composite optimization approach, integrating two different algorithmic 'materials' (SFLA and VNS) with complementary strengths. SFLA contributes its bio-inspired population dynamics and migration mechanisms, while VNS contributes its systematic neighborhood exploration strategies. This composite approach achieves both high convergence accuracy and robust scheduling rationality that neither algorithm could achieve alone.
2Device complexity
If traditional batch scheduling methods are used, then the scheduling process is simple, but job deterioration is not considered leading to unreasonable scheduling
Solution Approach 1:
The patent incorporates job deterioration by introducing time-dependent processing time parameters that change based on the start time of job execution. The processing time is modeled as a function of the current time, allowing the scheduling algorithm to account for deterioration effects. This parameter change enables the algorithm to generate more rational schedules that consider the dynamic nature of processing times, improving scheduling reliability without excessive complexity.
3Ease of operation
If the local search in shuffled frog leaping algorithm is similar to particle migration strategy, then the algorithm is easy to implement, but the discrete combinatorial optimization problem cannot be effectively solved
Solution Approach 1:
The patent introduces dynamic neighborhood structures from variable neighborhood search that adapt during the optimization process. The algorithm dynamically switches between different neighborhood structures (N1, N2, N3) based on search progress and effectiveness. This dynamic approach maintains ease of implementation through structured procedures while significantly enhancing optimization capacity for discrete combinatorial problems by exploring diverse solution spaces more effectively.
Data Source
AI summary
A method and a system for scheduling parallel machines based on hybrid shuffled frog leaping algorithm and variable neighborhood search algorithm are provided to solve collaborative production and processing of jobs on a plurality of unrelated batch processing machines. The jobs are distributed to machines based on the normal processing time and deterioration situation of the jobs on different machines and are arranged. An effective multi-machine heuristic rule is designed according to the structural properties of an optimal solution for the single-machine problem, and the improved rule is applied to the improved shuffled frog leaping algorithm to solve this problem. The improvement strategy for the traditional shuffled frog leaping algorithm is to improve the local search procedure of the traditional frog leaping algorithm by introducing the variable neighborhood search algorithm. The convergence rate and optimization capacity of the original algorithm are thus improved.