Development of an Analysis Platform Capable of Real-time Processing of Tens of Thousands of IoT Devices

Overview

The Business Analysis Center of Osaka Gas is working on developing a real-time analysis platform that analyzes and processes data sent from the company’s tens of thousands of IoT devices in real time.

Purpose

Due to the rapid increase in the amount of data resulting from the increase in the number of IoT devices managed by Osaka Gas, it is becoming difficult to analyze data in the existing environment created by the company. Moreover, in recent years, there has been a demand for systems that require real-time data analysis, machine learning, and control computation, as typified by virtual power plants, so it was necessary to develop a real-time IoT analysis platform.

System Configuration

Concept

This system not only collects and processes data in real time but also automatically performs machine learning. In addition, efforts are under way to automate accuracy monitoring and re-learning.
It is likely that the result will be a fully automated real-time analysis platform operated with much fewer staff, allowing the company’s IoT services to be expanded and enhanced.

Action

We participated in the Ministry of Economy, Trade and Industry’s Demonstration Project for Building Virtual Power Plant (FY2021) as a resource aggregator, controlled about 3,600 units of Ene-Farm, and established technology that could be active in actual business settings.