Predicting Overflow: A Novel Application of Latrine Sensors and Machine Learning for Optimizing Sanitation Services in Informal Settlements

• Activity Number: 341 - SPEED: Classification and Data Science
• Type: Contributed
• Date/Time: Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Learning and Data Science
• Author(s): Phillip Turman-Bryant* and Evan Thomas
• Companies: Portland State University and Portland State University
• Keywords: machine learning; sanitation; PLUMs; sensors; Kenya; Super Learner

Abstract

Servicing latrines in informal settlements can be expensive and difficult to optimize given the geospatial and temporal variability of latrine use. Daily servicing to avoid overflow events is inefficient, while dynamic scheduling of latrine servicing could reduce costs by providing just-in-time servicing for latrines. This study used motion detection sensors with cellular reporting and machine learning to dynamically predict when latrines could be skipped with a low risk of overflow. Sensors monitored daily latrine activity, and enumerators collected data related to solid and liquid waste weights and latrine servicing events. Given the complex relationship between latrine use and the need for servicing, an ensemble machine learning algorithm (Super Learner) was used to estimate waste weights and predict overflow events to facilitate dynamic scheduling. Accuracy of waste weight predictions based on sensor and historical weight data was adequate (mean absolute percent error of 20% and 22% for solid and liquid wastes), but there was greater accuracy in predicting overflow events with dynamic scheduling (area under the receiver operating characteristic curve of 0.90).