Optimizing Delivery Rider Allocation using Genetic Algorithm: A Demand Driven Approach for Efficient Order Fulfillment and Fair Earnings

Abstract

This thesis addresses the critical challenge of optimizing delivery rider allocation in Qatar’s rapidly expanding food delivery sector through an innovative genetic algo- rithm (GA) approach. The study develops a demand-driven multi-objective optimiza- tion model that simultaneously maximizes operational efficiency while ensuring fair earnings for gig workers—a persistent pain point in platform-based delivery systems. The proposed framework tackles four competing objectives: (1) minimizing operational costs, (2) maximizing order fulfillment rates, (3) optimizing rider utilization, and (4) guaranteeing minimum weekly earnings. Using a weighted fitness function with penalty constraints, the GA-based solution dynamically allocates riders across five geographic zones, adapting to fluctuating demand patterns. Key parameters include rider capacity, cost per order, and earnings per order, derived from real-world operational data. Through comparative analysis across five test scenarios—including fleet shortages and promotional demand surges—the model demonstrates superior performance to manual allocation methods, achieving about 90% utilization rates and 100% order fulfillment. Moreover, computational efficiency tests reveal consistent speed improvement, process- ing 7,000 rider allocations in 3.5 minutes versus 12 hours for manual methods. Sensi- tivity analysis further identifies optimal operational thresholds and trade-offs between workforce size, earnings, and service quality. The research makes three key contributions: (1) a GA solution approach for fair rider allocation, (2) empirical validation of the proposed approach for a case study based in Qatar, and (3) insights for platform operators on balancing efficiency with labor equity. While the model currently uses historical demand data, it can be integrated into dynamic environments.

Date of Award	2025
Original language	American English
Awarding Institution	HBKU College of Science and Engineering