Level: Intermediate
【Presenter】
Akira Hososhima, Representative of Tobayashi Consulting
[Presentation summary]
So far, we have introduced examples of using JMP to solve quality problems, such as reliability prediction methods, defect cause investigation methods, and quality trend analysis using text mining. This time, we used average analysis of proportions (ANOM) to extract and narrow down the model numbers and equipment with a significantly high number of defects from production data, and modeled the setting know-how of workers in a virtual experiment using a deterministic screening plan. Here's an example. When analyzing data that is added regularly, such as production data, it is easier to update by preparing fixed data such as product model number master in a separate table and linking the two. We will also introduce an example of an initiative in which we were able to reduce mistakes to 1/10 per year by visualizing human errors. We created a graph of the total number of operational errors and a breakdown of the errors, and used graphs stratified by individual, position, and organization to provide feedback to supervisors and workers to encourage awareness and ingenuity. We will introduce how to drop stratification variables into group X, columns, overlapping, etc. in the graph builder, and how to stratify using local data filters (LDF).
[Presenter profile]
After working at Yamatake Honeywell (currently Azbil) as a director, director of the R&D division, and director of the quality assurance promotion division, he founded Tobayashi Consulting. His areas of expertise include quality improvement, reliability prediction, robust design, general statistical problem solving such as experimental planning, design review, root cause analysis (RCA), and human error prevention.
His major publications are ``Practical Venture Business Success Strategies'' (Chuokeizaisha, 2011) and ``An Easy-to-Understand Book on Problem Solving'' (Nikkan Kogyo Shimbun, 2014) (single author). The main paper is ``Proposal for a workshop to promote the dissemination and reception of awareness regarding near misses and discomfort on the production line'' [2016 Quality Control Society Quality Technology Award Winner]. Main lectures include "Proposal of a mechanism to visualize organizational factors that cause work errors and promote improvement" (Discovery-Japan 2018), "Solving quality problems using JMP - Case study of failure analysis and reliability prediction" (Discovery-Japan 2018) 2019), "Improving the efficiency of experiments and analysis using JMP" (2020), "How to use text mining to investigate quality issues in high-mix, low-volume production lines" (2021), "Improving the efficiency of development, production, and quality assurance using JMP" Reform” 1st to 5th (JMP On Air Japanese version 2021)
