# Fighting the E-commerce Giants: Efficient Routing and Effective Consolidation for Local Delivery Platforms

**Authors:** Albert H. Schrotenboer, Michiel A. J. Uit het Broek, Paul Buijs, Marlin W. Ulmer  
**arXiv:** [2108.12608](https://arxiv.org/abs/2108.12608)  
**Date:** 2021-08-28  
**Summary page:** https://www.summarizepaper.com/en/paper/2108.12608v1/

## Key points

- Local delivery platforms offer instant-delivery from local stores to customers who order online
- Consolidation of orders is necessary for efficient and reliable delivery operations
- Delivery services cost-efficiently and reliably are a significant challenge for these platforms due to the complex, dynamic, stochastic pickup-and-delivery problem they face
- The paper introduces a generic approach that balances the consolidation potential and delivery urgency of orders using parameterized Cost-Function Approximation (CFA) approach
- The CFA approach modifies a set-packing formulation with two parameters to anticipate future demand while utilizing column generation to search the large decision space related to pickup-and-delivery problems quickly
- Numerical experiments show that the CFA approach strongly increases perceived customer satisfaction while lowering total travel time compared to various benchmark policies and reduces the percentage of late deliveries and their lateness to a minimum
- The proposed solution approach routes each product independently, bearing resemblance with static one-to-one pickup and delivery problems where efficient exact solution methods are readily available.
- The paper presents an innovative CFA approach that balances consolidation potential and delivery urgency while anticipating future demand in local delivery platforms' complex environment.

## Summary

Local delivery platforms are collaborative efforts where local stores offer instant-delivery to customers who order their products online. To increase the efficiency of the delivery operations and enable a high service guarantee towards both customers and stores, orders need to be consolidated. However, offering such delivery services cost-efficiently and reliably is a significant challenge for these platforms as they face a complex, dynamic, stochastic pickup-and-delivery problem. This paper introduces a generic approach that balances the consolidation potential and delivery urgency of orders. The newly developed parameterized Cost-Function Approximation (CFA) approach modifies a set-packing formulation with two parameters to anticipate future demand while utilizing column generation to search the large decision space related to pickup-and-delivery problems quickly. Inspired by an application in Groningen, Netherlands, numerical experiments show that our CFA approach strongly increases perceived customer satisfaction while lowering total travel time compared to various benchmark policies. Furthermore, it reduces the percentage of late deliveries and their lateness to a minimum. Our approach can assist managers in managing the non-trivial balance between consolidation opportunity and delivery urgency. The literature review identifies state-of-the-art approaches for efficient routing in deterministic pickup-and-delivery problems and outlines how dynamic routing is done in instant deliveries. The proposed solution approach routes each product independently, bearing resemblance with static one-to-one pickup and delivery problems where efficient exact solution methods are readily available. In conclusion, this paper presents an innovative CFA approach that balances consolidation potential and delivery urgency while anticipating future demand in local delivery platforms' complex environment. It provides valuable insights into improving customer satisfaction while reducing travel time and late deliveries' percentage through efficient routing techniques.