Accelerating Fleet Upgrade Decisions with Machine-Learning Enhanced Optimization

Sep 17, 2026

Uploaded by Noah

Rental-based business models and sustainability requirements increase the need for efficient strategies to manage large machine and vehicle fleet renewal and upgrades. Optimized fleet upgrade planning improves utility, cost, and sustainability, yet conventional integer-programming formulations scale exponentially and become computationally prohibitive. The study proposes an extended integer programming approach and a machine-learning method that reformulates the task as mixed discrete-continuous optimization, evaluated in an automotive case. Results show near-optimal solutions with strong scalability and computational performance.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

1. Introduction

[Available online at](Available online at www.sciencedirect.com)[ www.sciencedirect.com](Available online at www.sciencedirect.com) ScienceDirect Procedia CIRP 139 (2026) 7–12 13th CIRP Global Web Conference (CIRPe 2025) Accelerating Fleet Upgrade Decisions with Machine-Learning Enhanced Optimization Kenrick Howin Chaia, *, Stefan Hildebranda, *, Tobias Lachnitb, Martin Benferb, Gisela Lanzab, Sandra Klingea aDepartment of Structural and Computational Mechanics, TU Berlin, Straße des 17. Juni 135, Berlin and 10623, Germany bwbk Institute of Production Science, Karlsruhe Institute of Technology, Kaiserstraße 12, Karlsruhe and 76131, Germany * Corresponding author. Tel.: +49 (0) 30 314 21481; E-mail address: stefan.hildebrand@tu-berlin.de, [kenrick.howin.chai@campus.tu-berlin.de](kenrick.howin.chai@campus.tu-berlin.de) Abstract Rental-based business models and increasing sustainability requirements intensify the need for efficient strategies to manage large machine and vehicle fleet renewal and upgrades. Optimized fleet upgrade strategies maximize overall utility, cost, and sustainability. However, conventional fleet optimization does not account for upgrade options and is based on integer programming with exponential runtime scaling, which leads to substantial computational cost when dealing with large fleets and repeated decision-making processes. This contribution firstly suggests an extended integer programming approach that determines optimal renewal and upgrade decisions. The computational burden is addressed by a second, alternative machine learning-based method that transforms the task to a mixed discrete-continuous optimization problem. Both approaches are evaluated in a real-world automotive industry case study, which shows that the machine learning approach achieves near-optimal solutions with significant improvements in the scalability and overall computational performance, thus making it a practical alternative for largescale fleet management. © 2025 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ([https://creativecommons.org/licenses/by-nc-nd/4.0](https://creativecommons.org/licenses/by-nc-nd/4.0)) Peer review under the responsibility of the scientific committee of the CIRPe 2025 Keywords: Product upgrades; Replacement problem; Optimization; Machine learning; Integer programming 1. Introduction With rising sustainability demands and the need for longterm cost-effective operations, extending the lifetime of capital equipment has become increasingly important. Rather than relying on costly and recurring replacements, companies are increasingly exploring individualized upgrades and adaptations to maintain or enhance fleet performance over time. In ProductService Systems (PSS), enabling mid-life upgrades provides a promising strategy to reduce resource consumption, improve operational efficiency, and ensure continuous technological adaptation [1] . In fleet management, decisions still often focus on full replacements, although targeted upgrades represent a valuable alternative or addition to conventional renewal strategies. However, planning such upgrades at the fleet level introduces substantial complexity for manufacturers. Varying usage patterns, system aging, and heterogeneous product states must be considered while maintaining economic viability and scalability. Due to the scale of product fleets and the multitude of upgrade possibilities, conventional optimization approaches like mixed-integer programming (MIP) often face significant computational challenges. To address this challenge, the present study explores how machine learning (ML) can accelerate decision-making in fleet renewal scenarios by enabling the efficient planning and control of individualized product upgrades and adaptations at scale. The conventional Fleet Renewal Problem (FRP), also referred to as fleet replacement problem [2] asks for the optimal timing and

FAQs

Why are fleet upgrade decisions becoming more challenging in rental-based and sustainability-focused settings?

Rental-based models and sustainability requirements increase the need for efficient renewal and upgrade strategies. Handling large fleets and repeated decision-making makes conventional optimization computationally burdensome.

What two solution approaches are proposed in the study?

The study proposes an extended integer programming approach to determine renewal and upgrade decisions. It also introduces an alternative machine-learning-based method that reformulates the task into mixed discrete-continuous optimization.

What do the evaluation results show in the automotive industry case study?

The machine learning approach achieves near-optimal solutions while significantly improving scalability and overall computational performance. This makes it a practical alternative for large-scale fleet management.

Sep 17, 2026

Uploaded by Noah

Accelerating Fleet Upgrade Decisions with Machine-Learning Enhanced Optimization_Page 1

Open in WPS to view all 6 pages