1304. Bilevel Mixed-Integer Programming Model for the Network Design of a Recycling Waste Collection: A Case Study in a Chilean Commune
Invited abstract in session TB-4: Topics in Competition and Games, cluster Location, Network Design, and Routing.
Tuesday, 10:30-12:00Room: CE-209
Authors (first author is the speaker)
| 1. | Sebastián Dávila
|
| Industrial Engineering, Universidad de Santiago de Chile | |
| 2. | Marcela López G.
|
| Industrial Engineering , University of Santiago of Chile | |
| 3. | Franco Quezada
|
| Industrial Engineering Department, University of Santiago of Chile | |
| 4. | Sebastián San Martin C.
|
| Industrial Engineering , University of Santiago of Chile |
Abstract
Growing urban populations have led to increased waste generation, and authorities have been taking measures to reduce the amount of waste sent to landfills without being recycled, harming the environment and raising treatment costs. To address these challenges, authorities worldwide have proposed various strategies to encourage recycling, such as distributing recycling waste collection bins throughout the city for the population to deposit their recyclable materials (e.g. glass, cardboard, and plastic) for transportation to treatment centers, where they are converted into raw materials for new use.
To ensure optimal implementation of recycling waste collection bins, authorities must strategically determine the location, size, and bin category (glass, cardboard or plastic), based on user preferences. They must also define the collection frequency, to prevent micro-dumps near the bin due to exceeding capacity, and the routing to reduce logistical costs, including determining the model and size of collection trucks. Additionally, user decisions must be incorporated into the model to study their responses to incentives and/or costs associated with dropping off materials in the recycling collection bin.
This paper optimizes the location of recycling waste collection bins to increase recycling and reduce transportation costs. We mathematically formulate the problem as a bilevel mixed-integer programming (BMIP) model based on the periodic location-routing-inventory problem with users' preferences.
We present a case study of a commune in Chile to test the formulation and evaluate various decision-making strategies, including economic costs, penalty levels for lost users and excess waste, buying or leasing collection trucks, and changes in transportation costs and demand levels.
Keywords
- Location and Routing
- Sustainable Development
- Bilevel programming
Status: accepted
Back to the list of papers