Home > Research > Publications & Outputs > Model formulations for pickup and delivery prob...

Links

Text available via DOI:

View graph of relations

Model formulations for pickup and delivery problems in designated driver services

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Model formulations for pickup and delivery problems in designated driver services. / Arslan, Alp; Agatz, Niels; Srour, Jordan.
In: Computers and Operations Research, Vol. 164, 106547, 30.04.2024.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Arslan, A., Agatz, N., & Srour, J. (2024). Model formulations for pickup and delivery problems in designated driver services. Computers and Operations Research, 164, Article 106547. https://doi.org/10.1016/j.cor.2024.106547

Vancouver

Arslan A, Agatz N, Srour J. Model formulations for pickup and delivery problems in designated driver services. Computers and Operations Research. 2024 Apr 30;164:106547. Epub 2024 Feb 2. doi: 10.1016/j.cor.2024.106547

Author

Arslan, Alp ; Agatz, Niels ; Srour, Jordan. / Model formulations for pickup and delivery problems in designated driver services. In: Computers and Operations Research. 2024 ; Vol. 164.

Bibtex

@article{04551c8620a54dc3ae9ead6588e08c81,
title = "Model formulations for pickup and delivery problems in designated driver services",
abstract = "Designated driver services use company vehicles to deliver drivers to customers. The drivers then drive the customers from their origins to their destinations in the customers{\textquoteright} own cars; at the destinations, the drivers are picked up by a company vehicle. We typically see teams of drivers assigned to company vehicles serving customers. However, when the drivers may be dropped off by one vehicle and picked up by another, a challenging pick-up and delivery problem arises. In this paper, we study the structural properties of the designated driver problem focusing on the synchronization between company vehicles and drivers. We present a two-index formulations to generate optimal, least-cost routes using a general-purpose solver. We benchmark the two-index formulations against a 3-index formulation and a path enumeration strategy. Based on a set of experiments, we find that the two-index formulation performs well, both in terms of quality and solution time, especially on the formulations with more flexibility in the pairing of drivers to company vehicles. Our computational experiments show that up to 75% cost savings are possible from using a flexible operating strategy as compared to a strategy in which drivers and company vehicles stay together throughout a shift.",
keywords = "Routing, Dial-a-ride, Pickup and delivery problem, Designated driver services, Routing with precedence",
author = "Alp Arslan and Niels Agatz and Jordan Srour",
year = "2024",
month = apr,
day = "30",
doi = "10.1016/j.cor.2024.106547",
language = "English",
volume = "164",
journal = "Computers and Operations Research",
issn = "0305-0548",
publisher = "Elsevier Ltd",

}

RIS

TY - JOUR

T1 - Model formulations for pickup and delivery problems in designated driver services

AU - Arslan, Alp

AU - Agatz, Niels

AU - Srour, Jordan

PY - 2024/4/30

Y1 - 2024/4/30

N2 - Designated driver services use company vehicles to deliver drivers to customers. The drivers then drive the customers from their origins to their destinations in the customers’ own cars; at the destinations, the drivers are picked up by a company vehicle. We typically see teams of drivers assigned to company vehicles serving customers. However, when the drivers may be dropped off by one vehicle and picked up by another, a challenging pick-up and delivery problem arises. In this paper, we study the structural properties of the designated driver problem focusing on the synchronization between company vehicles and drivers. We present a two-index formulations to generate optimal, least-cost routes using a general-purpose solver. We benchmark the two-index formulations against a 3-index formulation and a path enumeration strategy. Based on a set of experiments, we find that the two-index formulation performs well, both in terms of quality and solution time, especially on the formulations with more flexibility in the pairing of drivers to company vehicles. Our computational experiments show that up to 75% cost savings are possible from using a flexible operating strategy as compared to a strategy in which drivers and company vehicles stay together throughout a shift.

AB - Designated driver services use company vehicles to deliver drivers to customers. The drivers then drive the customers from their origins to their destinations in the customers’ own cars; at the destinations, the drivers are picked up by a company vehicle. We typically see teams of drivers assigned to company vehicles serving customers. However, when the drivers may be dropped off by one vehicle and picked up by another, a challenging pick-up and delivery problem arises. In this paper, we study the structural properties of the designated driver problem focusing on the synchronization between company vehicles and drivers. We present a two-index formulations to generate optimal, least-cost routes using a general-purpose solver. We benchmark the two-index formulations against a 3-index formulation and a path enumeration strategy. Based on a set of experiments, we find that the two-index formulation performs well, both in terms of quality and solution time, especially on the formulations with more flexibility in the pairing of drivers to company vehicles. Our computational experiments show that up to 75% cost savings are possible from using a flexible operating strategy as compared to a strategy in which drivers and company vehicles stay together throughout a shift.

KW - Routing

KW - Dial-a-ride

KW - Pickup and delivery problem

KW - Designated driver services

KW - Routing with precedence

U2 - 10.1016/j.cor.2024.106547

DO - 10.1016/j.cor.2024.106547

M3 - Journal article

VL - 164

JO - Computers and Operations Research

JF - Computers and Operations Research

SN - 0305-0548

M1 - 106547

ER -