[Paper Review] Robust Order Scheduling in the Fashion Industry: A Multi-Objective Optimization Approach

Wei Du, Yang Tang+4 more|arXiv (Cornell University)|Feb 1, 2017

Scheduling and Optimization Algorithms17 references4 citations

TL;DR

This paper proposes a robust multi-objective order scheduling approach for the fashion industry that integrates pre-production events and uncertainties in daily production quantities using the NSJADE evolutionary algorithm. It demonstrates that considering pre-production events and production variability significantly improves schedule robustness and reduces tardiness, with NSJADE outperforming standard JADE and non-robust variants in minimizing total clashes and tardiness under uncertainty.

ABSTRACT

In the fashion industry, order scheduling focuses on the assignment of production orders to appropriate production lines. In reality, before a new order can be put into production, a series of activities known as pre-production events need to be completed. In addition, in real production process, owing to various uncertainties, the daily production quantity of each order is not always as expected. In this research, by considering the pre-production events and the uncertainties in the daily production quantity, robust order scheduling problems in the fashion industry are investigated with the aid of a multi-objective evolutionary algorithm (MOEA) called nondominated sorting adaptive differential evolution (NSJADE). The experimental results illustrate that it is of paramount importance to consider pre-production events in order scheduling problems in the fashion industry. We also unveil that the existence of the uncertainties in the daily production quantity heavily affects the order scheduling.

Motivation & Objective

• To address the gap in fashion order scheduling research by incorporating pre-production events, which are often overlooked despite their real-world impact.
• To develop a robust order scheduling framework that minimizes sensitivity to daily production quantity variations during actual production.
• To optimize three objectives: minimizing total pre-production event clashes, minimizing total order tardiness, and enhancing schedule robustness against production variability.
• To validate the effectiveness of robust multi-objective optimization in improving real-world scheduling performance in fashion manufacturing.
• To provide a novel, practical scheduling solution that accounts for both collaborative pre-production processes and operational uncertainties.

Proposed method

• The problem is formulated as a multi-objective optimization model with three objectives: minimizing pre-production event clashes, minimizing total tardiness, and maximizing robustness to production quantity uncertainty.
• A robust multi-objective evolutionary algorithm, nondominated sorting adaptive differential evolution (NSJADE), is employed to solve the scheduling problem.
• NSJADE incorporates adaptive mutation and crossover strategies to enhance convergence and diversity in the Pareto front search.
• Uncertainty in daily production quantities is modeled as interval-based variations, and robustness is evaluated through worst-case performance across uncertain scenarios.
• The algorithm generates a set of non-dominated solutions that balance the three objectives, enabling decision-makers to select robust, practical schedules.
• The approach is validated using a real-world-inspired case study based on Fast React software data, simulating industrial scheduling complexity.

Experimental results

Research questions

• RQ1 How do pre-production events affect the feasibility and quality of order schedules in the fashion industry?
• RQ2 To what extent does uncertainty in daily production quantities degrade the performance of conventional order schedules?
• RQ3 Can a robust multi-objective optimization approach significantly improve schedule stability and reduce tardiness under production variability?
• RQ4 How does NSJADE compare to standard JADE and non-robust NSJADE in handling the trade-offs between pre-production clashes, tardiness, and robustness?
• RQ5 What is the impact of integrating pre-production events into the scheduling model on overall scheduling performance?

Key findings

• Considering pre-production events is critical for realistic and effective order scheduling, as their omission leads to infeasible or suboptimal schedules.
• The existence of uncertainty in daily production quantities significantly degrades the performance of non-robust schedules, increasing tardiness and schedule instability.
• NSJADE outperforms standard JADE and non-robust NSJADE in maintaining lower total tardiness and pre-production clashes across uncertain scenarios.
• The proposed robust multi-objective approach successfully balances three conflicting objectives: minimizing clashes, minimizing tardiness, and maximizing robustness.
• The experimental results confirm that robustness to production variability is a decisive factor in achieving reliable and practical schedules in real-world fashion manufacturing.
• This study is the first to integrate pre-production events and production uncertainty into a unified multi-objective scheduling framework for the fashion industry.