TY - UNPB

T1 - Predicting Energy Customer Vulnerability Using Smart Meter Data

AU - Ushakova, Anastasia

AU - Jankin MIkhaylov, Slava

PY - 2018

Y1 - 2018

N2 - Supporting vulnerable consumers and reducing fuel poverty are major priorities for policy makers in the energy sector. With the availability of streamingdata from smart meters we are able to develop simple and reliable methods ofidentifying vulnerable energy customers and as a result develop targeted policyinterventions. This study investigates how vulnerable customers can be identified from natural gas consumption data. Neural networks, random forest, naiveBayes, and support vector machines were assessed for classification of consumervulnerability. Random forest, with the prediction accuracy of 94.6 percent, outperforms other prediction models. Our study provides additional evidence thatmachine learning methods can be deployed by policymakers and insights teamsto predict vulnerability from patterns of consumer behaviour.

AB - Supporting vulnerable consumers and reducing fuel poverty are major priorities for policy makers in the energy sector. With the availability of streamingdata from smart meters we are able to develop simple and reliable methods ofidentifying vulnerable energy customers and as a result develop targeted policyinterventions. This study investigates how vulnerable customers can be identified from natural gas consumption data. Neural networks, random forest, naiveBayes, and support vector machines were assessed for classification of consumervulnerability. Random forest, with the prediction accuracy of 94.6 percent, outperforms other prediction models. Our study provides additional evidence thatmachine learning methods can be deployed by policymakers and insights teamsto predict vulnerability from patterns of consumer behaviour.

M3 - Working paper

BT - Predicting Energy Customer Vulnerability Using Smart Meter Data

ER -