Optimal patrol routing and scheduling for parking enforcement considering drivers' parking behavior

Liu, Xinlu

Permalink

https://hdl.handle.net/2142/78687 Copy

Description

Title

Optimal patrol routing and scheduling for parking enforcement considering drivers' parking behavior

Author(s)

Liu, Xinlu

Issue Date

2015-04-29

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• continuum approximation
• patrol routing

Abstract

Logistics costs constitute a considerable proportion of overall daily expenses for many public sectors, among which parking enforcement agencies are some of the most prominent examples. While currently there is little research about the planning of efficient parking enforcement patrol operations, this work presents several models to generate patrol schemes that help parking departments achieve low operational costs and effective enforcement. This thesis considers two levels of problems: i) parking behavior of drivers based on given patrol frequency (but not schedule), and ii) parking enforcement patrol routing and scheduling based on the parking behavior of drivers. Driver determines optimal payment based on the distribution of parking duration, parking prices, citation fines, and patrol frequencies via a newsvendor model. As the intensity of parking enforcement increases, illegal parking is expected to occur less frequently. However, improving parking enforcement sometimes requires more frequent patrols that lead to higher agency costs. In order to find the optimal trade-off point, the problem is further formulated into a Vehicle Routing Problem (VRP). Solving this bi-level optimization problem means that the cost is reduced while anticipated parking offenses are limited to a certain level. We present a traditional discrete mixed-integer programming model, and a continuous approximation model based on the method of continuum approximation. Numerical tests are performed in order to examine the performance of these two models using randomly-generated datasets. Sensitivity analyses show that as parking price or demand increases, or citation fine decreases, more frequent patrols are required to maintain the healthy operation of the parking lots. The results also validate that the method of continuum approximation can offer good estimation of the agency cost for the parking patrol problem with comparatively minimal runtime.

Graduation Semester

2015-5

Type of Resource

text

Permalink

http://hdl.handle.net/2142/78687

Copyright and License Information

Copyright 2015 Xinlu Liu

Owning Collections