An Integrated Batch Scheduling Model for Sequential Flow Shops in a Production Network Processing a Single Item to Minimize Total Actual Flow Time

Nasution, Andri Rachmat Kumalasian; Isnaini, Mohammad Mi'radj; Suprayogi, Suprayogi; Halim, Abdul Hakim

doi:10.22068/ijiepr.37.3.2800

Volume 37, Issue 3 (IJIEPR- 2026) IJIEPR 2026, 37(3): 240-266 | Back to browse issues page

‎ 10.22068/ijiepr.37.3.2800

Mendeley

Zotero

RefWorks

Nasution A R K, Isnaini M M, Suprayogi S, Halim A H. An Integrated Batch Scheduling Model for Sequential Flow Shops in a Production Network Processing a Single Item to Minimize Total Actual Flow Time. , IJIEPR. 2026; 37 (3) :240-266
URL: http://ijiepr.iust.ac.ir/article-1-2800-en.html

An Integrated Batch Scheduling Model for Sequential Flow Shops in a Production Network Processing a Single Item to Minimize Total Actual Flow Time

Andri Rachmat Kumalasian Nasution¹

, Mohammad Mi'radj Isnaini¹

, Suprayogi Suprayogi¹

, Abdul Hakim Halim ^*²

1- Faculty of Industrial Technology, Institut Teknologi Bandung, Jl. Ganesha No. 10, Bandung 40132, Indonesia
2- Faculty of Industrial Technology, Institut Teknologi Bandung, Jl. Ganesha No. 10, Bandung 40132, Indonesia , ahakimhalim@itb.ac.id

Abstract: (454 Views)

This study develops an integrated batch-scheduling model for sequential flow shops in a production network processing a single item under a common due date. The model simultaneously determines the number of batches, batch sizes, and a feasible backward schedule to minimize the Total Actual Flow Time while accounting for stage-specific processing and setup times. An independent Gurobi procedure evaluates every integer value for the number of batches in the entire computational domain and solves the corresponding nonconvex model via spatial branch-and-bound with a zero-gap target, subject to solver numerical tolerances. A heuristic is developed to reduce this computational effort by combining a continuous estimate of the number of batches, KKT-based batch sizing, backward-schedule reconstruction, and directional neighborhood search. The computational experiment compares the proposed heuristic with the independently obtained Gurobi reference and a Simulated Annealing benchmark. All 300 benchmark instances are globally resolved by Gurobi over the complete computational domain under the final solver protocol, and the proposed heuristic matches the reported Gurobi objective and selected number of batches at the adopted reporting precision. Simulated Annealing provides an additional independent stochastic comparison. Sensitivity analyses examine demand, processing times, production-network length, and minimum allowable batch size, and a real-data application demonstrates operational relevance under explicitly stated practical limitations.

Keywords: Production Network, Flow Shop, Batch Scheduling, Heuristic Algorithm, Total Actual Flow Time, Backward Scheduling

Full-Text [PDF 1007 kb] (202 Downloads)

Type of Study: Research | Subject: Operations Research
Received: 2026/06/21 | Accepted: 2026/09/5 | Published: 2026/09/8