EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
2359. Estimating Urban Solid Waste Generation for Collaborative Collection Routes using Spatial Analysis and Machine Learning
Invited abstract in session MB-18: Assessment Methods for Shaping the Green, Inclusive, and Digital Cities II, stream Sustainable Cities.
Monday, 10:30-12:00Room: 42 (building: 116)
Authors (first author is the speaker)
| 1. | Carola Blazquez
|
| Department of Engineering Science, Universidad Andres Bello | |
| 2. | Victor Silva
|
| Ciencias de la Ingenieria, Universidad Andrés Bello |
Abstract
In the last decades, there has been an increasing generation of solid waste in urban areas, imposing an environmental, economic, and social burden worldwide. Thus, there is a need to estimate solid waste generation rates for obtaining efficient collection routes and itineraries, particularly when two or more communes decide to collaborate their resources (e.g., vehicles and staff), in order to reduce costs and CO2 emissions, and increase equitable workload. This study performs spatial analysis using census data, and historical data based on daily GPS measurements and tare weights (tonnage) of collection vehicles in two communes in Santiago, Chile to identify collection zones and estimate initial solid waste generation rates at the street level. Machine learning techniques are applied to predict daily solid waste generation rates using demographic characteristics, land-use, socioeconomic status, and weather information, which are subsequently tested with the initial solid waste generation rates estimations. Prediction results are promising when compared with these initial estimations. Our results provide important insights for supporting the optimization of collaborative collection truck routes among different communes in urban areas.
Keywords
- Machine Learning
- Vehicle Routing
- Sustainable Development
Status: accepted
Back to the list of papers