The introduction of a big data platform for ICT-based large volume production data integration is accelerating in a situation where the complexity of recent manufacturing processes has increased and a high level of efficiency is required.
Meanwhile, various software platforms for smart factory have been proposed. These kinds of automated systems for the manufacturing industry have been being implemented, each of which is a fusion of ICT and manufacturing industry, and manufacturing processes can also be optimized in Industry 4.0.
In order to implement a smart factory, it is necessary to have an integrated platform that is optimized the automation process to match with process Life-Cycle. This is also a fusion of IoT(Internet of Things)-based CPS(Cyber Physical System) technology and ICT technology.
In case of large enterprises, ICTs are actively utilized, but small and medium-sized manufacturers are difficult to introduce systems due to problems of industrial structure and factors of vulnerability that small and medium-sized manufacturers themselves have.
In this paper, we provide an application method of 4M data-based big data platform and analysis which can be flexibly applied to the application considering the extensibility necessary for smart factory implementation of small and medium manufacturing enterprises.

1. Manufacturing data in small and medium manufacturing enterprises in automobile parts industry was collected, classified into a 4M (Man, Machine, Material, Method) data, and stored in a database (an integrated operational data store).
2. We built a cloud-based big data analysis system for large volume data processing. Big data analysis system has been implemented by using open source SWs for making low cost platform. 4M database is periodically stored in the Hadoop Ecosystem and can be analyzed using Spark R. The result can be published in the Web interface by using R Shiny library.
3. With the big data analysis system, we analyzed process pattern defects, 4M data analysis, reliable base equipment failure prediction, overall equipment efficiency analysis, and found 4M factors that affect productivity.

The results of the analysis are summarized as following.
As result of pattern analysis of the process defact status, defect it was found that there were many defects due to the worker factors. 4M data analysis result shown that there was a difference in defective quantity depending on workers, materials, working method.
In the overall equipment efficiency analysis, the productivity impact from equipment failure was small. It indicated that the decrease in productivity effect is large due to insufficient workers’ response to idling and momentary stopping of facilities.

Through 4M data analysis, we can quickly identify problems, obtain improvement strategy in production and quality control. Futhure more, it can also support process optimization and smart decision making.