Data Analysis Seminar for Engineers: Learning through "Try & Error"
~Practice of Factor Exploration and Predictive Model Building~
This seminar will focus on engineers in the manufacturing industry and will discuss data utilization, specifically practical approaches to factor exploration and predictive model building using JMP products. Analysis does not yield optimal results on the first attempt; it is crucial to enhance accuracy through repeated trial and error in data preparation and model building. In particular, the quality of data preparation, such as preprocessing and feature creation, significantly influences the results, and JMP allows for efficient progress while visualizing the process. Using publicly available data for model building, we will introduce a series of processes from preparation, including visualization, preprocessing, and feature creation, to factor exploration through correlation and regression analysis, predictive model building (such as neural networks and random forests), and extraction of important factors, all through trial and error. We will also showcase some lesser-known convenient features of JMP during the seminar. We encourage you to participate in this opportunity. Target Participants: - Engineers who want to conduct quality improvement and factor analysis based on data. - Those who want to apply the entire flow of predictive model building, including preprocessing and feature creation, in practice. - Individuals who wish to efficiently build and evaluate factor exploration and machine learning models using JMP products.
Inquire About This Product
basic information
Date and Time: December 15, 2026 (Tuesday) 14:00–15:30 (Online seminar using Zoom) Program: 1) What to do before modeling: Key points for data preparation and trial & error using JMP * Understanding the contents of the data (visualization and summary statistics) * Dealing with missing values and outliers * Creating variables to enhance model accuracy (feature engineering) * Tips for efficiently progressing with data preparation for analysis 2) How to explore factors and build predictive models without creating a black box * Visualization of factors using correlation analysis, regression models, and decision trees * Model evaluation using training and validation data * Comparison and selection of multiple machine learning models (e.g., neural networks, random forests, etc.) * Intuitive understanding of important factors and prediction results
Price information
Participation fee: Free (Pre-registration is required from the page below) https://www.jmp.com/ja/events/live-webinars/non-series/2026/12-15-factor-exploration
Delivery Time
Applications/Examples of results
"Democratization of Data" and Sharing the "Wisdom" of Manufacturing Across the Company - TOTO's New Exploration of "Good Products and Uniformity" [Challenge] The manufacturing of sanitary ceramics using natural raw materials involves approximately 13% shrinkage during the drying and firing processes. As products become larger and more complex, the challenge is how to maintain "uniform" high quality and convert the "tacit knowledge" of skilled artisans into "explicit knowledge" to pass on to the next generation. This has been a crucial issue at TOTO Ltd., which has a history of over 100 years and forms the foundation of the company's manufacturing. [Solution] An "exploratory data analysis" using JMP was introduced for the manufacturing data from the advanced Shiga factory. By implementing the first barcode system in the sanitary ceramics factory, we deepened the high yield achieved and quantified the "good product conditions" using visual verification with graph builders, as well as partitioning and cluster analysis, leading to improvements in direct yield and overall yield. Additionally, JMP was adopted for the foundational education of the "in-house study abroad program" over two years, aiming to enhance company-wide data science skills. [Results] Please check the following page! https://www.jmp.com/ja/customer-stories/toto
catalog（1）
Download All Catalogs
Company information
The JMP story goes back to 1989 when John Sall decided to combine statistical analysis capabilities with graphical visualizations to animate and visualize data. For more than 35 years now, John Sall has led JMP R&D, making each version of JMP more visual, more interactive, and more practical to help users understand their data. What started as a passion project has grown by leaps and bounds. It’s now a family of statistical software products designed with scientists and engineers in mind and used worldwide in nearly every industry.
