•  
  •  
 

Abstract

In a production system, rework plays a key role in reducing waste and efficiently managing manufacturing costs. The proposed model addresses a system with batch arrivals, a single server, and a two-stage bulk service with inspection, two rework cycles, and multiple vacations. First, batches are processed in stage 1, some of which may require rework during the same cycle, albeit with a certain probability, while others proceed to stage 2. After stage 2, batches are inspected, and if evidence of nonconformities is found, they are reworked with a certain probability. After inspection or rework, batches are processed according to the current queue length. If the queue length is equal to or greater than a specific limit (Q ≥ a), the server returns to serving the next batch; otherwise, the system enters multiple vacations, which help maximize server availability and minimize idle time. The rework management model is applied in the paint manufacturing industry to streamline rework operations and enhance manufacturing performance. The supplementary variable technique is used to build a model for studying the probability-generating function of the queue size. Performance measures are then illustrated with numerical examples, and the cost-effectiveness of the results is shown using graphs. The study presents noteworthy findings by comparing the neuro-fuzzy technique results against outcomes obtained through Adaptive Neuro-Fuzzy Inference System (ANFIS) validation.

Keywords

ANFIS, Bulk service, Sensitivity analysis, Single server, Two-stage, Two rework cycles

Subject Area

Mathematics

Article Type

Article

First Page

2922

Last Page

2941

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Download

Share

 
COinS