Research of Flexible Assembly Job-Shop Batch-Scheduling Problem Based on Improved Artificial Bee Colony

Area of Science:

• Operations Research
• Manufacturing Engineering
• Industrial Engineering

Background:

• The flexible assembly job-shop scheduling problem with lot streaming (FAJSP-LS) is crucial for multivariety, small-batch production.
• Traditional large-batch processing in FAJSP-LS leads to increased waiting times, particularly at the assembly stage.
• Batch splitting is explored as a strategy to mitigate these inefficiencies.

Purpose of the Study:

• To investigate the impact of batch splitting (unequal and consistent sizes) on the FAJSP-LS.
• To develop an efficient algorithm for solving the integrated batch splitting and scheduling problem.
• To enhance time efficiency in two-stage job-shop systems.

Main Methods:

• The problem was formulated as a mixed-integer linear program, separating batch splitting and scheduling.
• An improved bio-inspired algorithm, based on the artificial bee colony, was developed.
• A four-layer chromosome encoding and a multi-colony optimization strategy were employed.

Main Results:

• The proposed algorithm demonstrated superior solution quality compared to non-split or equally split batches in benchmark tests.
• Application to a refrigerator workshop confirmed improved time efficiency with unequal batch splitting.
• The algorithm effectively addresses the NP-hard nature of the integrated scheduling problem.

Conclusions:

• Splitting batches, especially into unequal sizes, is a viable strategy to improve FAJSP-LS performance.
• The enhanced artificial bee colony algorithm provides an effective solution for complex scheduling and batching problems.
• Unequal batch splitting offers practical advantages for time efficiency in manufacturing settings.