Machine Learning For Determining Planned Order Lead Times In Job Shop Production - A Systematic Review Of Input Factors And Applied Methods

Aug 4, 2026

Upload by Skyler

Accurate planned order lead time determination helps firms confirm feasible delivery dates while improving production capacity and procurement planning. Common industry approaches often rely on simplified assumptions and limited input-factor coverage, making them difficult to adapt when workloads shift or staffing shortages occur. For make-to-order job shop production, late predictions cause delayed operations and weaker delivery performance, while early predictions increase unnecessary inventory. This study delivers a systematic literature review identifying relevant input factors, the machine learning methods used to quantify them, and the applied feature selection, regression, evaluation metrics, and explainable AI techniques, enabling assessment of research gaps.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

1. Introduction

2. Theoretical Background

3. Methodology (Systematic Literature Review)

4. Review Results

Input factors

Models applied

Methods for identifying important input factors

Summary and Outlook

CONFERENCE ON PRODUCTION SYSTEMS AND LOGISTICS CPSL 2025 7th Conference on Production Systems and Logistics Machine Learning For Determining Planned Order Lead Times In Job Shop Production: A Systematic Review Of Input Factors And Applied Methods Ferenc Wolter 1, Alexander Rokoss 1, Matthias Schmidt2 1Institute for Production Technology and Systems (IPTS) / Leuphana University Lueneburg, Lueneburg, Germany 2Institute of Production Systems and Logistics (IFA) / Leibniz University Hanover, Hanover, Germany Abstract The accurate planning of order lead times enables companies to confirm feasible delivery times to their customers and facilitates more efficient planning of production capacities and procurement processes. In practice, the commonly used methods for determining planned order lead times are constrained by simplified assumptions and the limited consideration of input factors. As a result, they struggle to adapt to changing conditions, such as varying production workloads or short-term employee shortages. For manufacturers engaged in make-to-order production with complex structures, such as job shop production, consequently delayed operations result in an insufficient delivery performance. Conversely, early operations lead to the accumulation of unnecessary inventories. In this context, machine learning methods are expected to offer significant potential for dealing with changing circumstances due to their ability to utilize a wide range of input factors. A variety of machine learning approaches incorporating diverse data sets have been proposed in the literature. This paper presents the findings of a systematic literature review on the potential input factors for determining planned order lead times in job shop production. Moreover, the utilized machine learning methods to quantify these input factors are identified. For this purpose, the input data used in case studies, the machine learning methods applied for both feature selection and regression analysis, as well asthe evaluation metrics and explainable artificial intelligence approaches, are analyzed and synthesized. This allows the identification of research gaps regarding input factors and their quantification for determining planned order lead times. Keywords Planned Order Lead Time; Machine Learning; Job Shop Production; Input Factors; Prediction Methods; Production Planning And Control 1. Introduction In recent decades, customers' demands on the logistical performance of production companies rised steadily. In addition to high-quality products, customers today expect short and, in particular, reliable delivery times [1] . In production companies that adhere to the job shop principle, order lead time is typically the major component of delivery time due to the order-specific manufacturing processes [2] . The prediction of order lead times in job shop production has therefore been the subject of research for decades [3] . The use of standard order lead times for planning often results in insufficient prediction accuracy due to widely varying order contents [4,5]. If the order lead time predictions underestimate the required time, adherence to promised delivery times necessitates extensive interventions in order scheduling, which in turn results in DOI: [https://doi.org/10.15488/1888](https://doi.org/10.15488/1888)􀀕 ISSN: 2701-6277 374 prolonged order lead times for other orders. Conversely, order lead time predictions that overestimate the required time result in a reduction in the logistical performance perceived by customers and, thus, indirectly lead to negative effects on sales. The large number of input factors influencing order lead times in job shop production makes precise prediction even more difficult. As a result, stochastic, heuristic and model-based approaches have not been able to generate a satisfactory prediction quality in the past [3,6–8] . Machine learning (ML) represents a promising approach for predicting order

FAQs

Why are planned order lead times important in job shop production planning?

They enable feasible delivery promises and support more efficient production capacity planning and procurement processes. In job shops, order lead time strongly drives delivery time due to order-specific manufacturing processes.

What limitations exist in commonly used methods for determining planned order lead times?

They often use simplified assumptions and consider only a limited set of input factors. This reduces adaptability under changing conditions like workload variation or short-term employee shortages.

Which research questions does the systematic review address?

RQ1 identifies input factors used for ML-based order lead time prediction; RQ2 specifies ML regression models applied; RQ3 covers methods used to quantify how individual input factors affect prediction quality.

Aug 4, 2026

Upload by Skyler