CAMSL Dataset: Smart Meter Data under Time-Of-Use Tariffs in Japan

Published: 22-12-2020| Version 1 | DOI: 10.17632/cmpsyncmmk.1
Yohei Kiguchi,
Melvyn Weeks,
Riku Arakawa


CAMSL is the first public dataset for a Time-Of-Use (TOU) tariff intervention study using smart-meter data including pre, during and post TOU intervention periods. It includes 1423 households (1023 TOUusers and 400 Non-TOU users) in Tokyo between 1st July 2017 and 31st December 2018 (18 months). The dataset also includes raw data of 3337 customers who did not participate in the TOU trial. Each day has 48 half-hourly data points for energy consumption from a smart meter and each household has 579 days between 1 July 2017 to 31 December 2018, comprising a total of 27792 data points for electricity consumption obtained at each household for this dataset. The uniqueness of this dataset is the included online engagement data recorded via web-application usage, which enables further studies related to gamification effects. half-hourly consumption data from 1st June 2017 to 31st December 2018 customer_info.csv: customer information (house_type, number_of_residents, tou) TOU users == 1 Control users == 0 As for the selection of Control users, please refer to the article. web_info.csv: web activity information (for TOU customers) (sessions, average_session_duration, bounce_rate) already padded if the value is missing. temperature_Tokyo.csv: hourly temperature data in Tokyo from 1st June 2017 to 31st December 2018 holidays.csv: Japanese national holidays non_tou.csv.gz: raw data of consumption of total 3337 customers who did not participate in the TOU trial