IoT sensors can monitor manufacturing processes to ensure consistent quality and identify deviations from desired performance parameters. AI algorithms can analyze process data to optimize manufacturing parameters, reduce defects, and improve product performance. Machine learning models can identify correlations between process parameters, material properties, and product performance to guide process optimization efforts. Overall, the integration of IoT and AI technologies into tribology offers significant opportunities for improving reliability, efficiency, and performance across various industries, including manufacturing, automotive, aerospace, and energy. By leveraging real-time data analytics and predictive capabilities, organizations can optimize maintenance practices, minimize downtime, and enhance the longevity of critical components. This paper outlines mechanical maintenance based on lubricating oil and condition monitoring using IoT. I also describe our research on the application of AI to analytical ferrography as an example.

In recent times, the advancement of artificial intelligence (AI) has been significant, enabling its integration and the innovation of materials in various sectors. Tribology has seen a number of initiatives aimed at predicting friction and wear, although AIʼs adoption here lags behind its use in other areas. This paper highlights three main areas: the use of deep learning to determine the friction coefficient from visual representations of friction surfaces, the prediction of wear, and the development of a unique function that prevents interaction with damaged areas on opposing surfaces through the use of adaptable surfaces controlled by AI. When employing AI for these predictions, a common challenge among engineers and scientists is the interpretation of the outcomes. Consequently, the domain of AI that is understandable and interpretable has seen rapid growth. Nonetheless, for those not deeply versed in AI, staying abreast of its swift progress poses a challenge. This underscores the growing necessity for joint ventures with experts in AI to bridge this gap.

We are promoting digital transformation of the manufacturing industry for further strategic use of the latest digital technology and data. In order to strengthen existing businesses through the use of data, analysis of various types of business data is essential to promote this, and therefore the company is collecting and analyzing manufacturing data, product IoT data, and customer data. One example of this is by analyzing vehicle data from connected motorcycles on the market, it is possible to more accurately understand usersʼmarket levels, which was previously an unknown sector. In addition, by analyzing the userʼs activities records on the web page, it is further possible to understand the customerʼs journey leading up to the purchase of a product, etc. Using data and these measures, the company aims to create new, and greater customer experiences. This paper introduces the data analysis platform to conduct data analysis. Specifically, we will introduce a “data analysis platform” that enables data collection, conversion, analysis, reporting, and other processes to be performed by data stores and analysis sites. In addition, examples of data analysis for IoT products, marketing, and manufacturing systems using this data analysis platform will also be introduced.

Hydraulic fluid management, checking and maintaining oils, is absolutely paramount importance in order to keep a hydraulic system operating efficiently and effectively. Currently many companies waste a lot of time collecting samples and analyzing it in a laboratory. We would like to introduce a new hydraulic fluid monitoring system that checks and shows oil conditions on-site, and it can be prevented the troubles in advance and kept stable operation.

Condition monitoring is being required more and more from the customer's needs to reduce maintenance cost and time, downtime of the vehicles and sudden accidents. As the technology to prevent them, monitoring technology of bypass valve stroke sensors which can detect whether the filter for hydraulic oil is clogged and rotation feedback amount of hydraulic fan pump and motor which can detect efficiency decrease of hydraulic fan system for oil coolers are introduced. As a result of the investigations vehicles with these technologies, it is cleared both of them can detect tendency of failure in the field. Completely clogged filter and damaged parts of hydraulic piston pump were found by high frequency of abnormal signs from bypass valve stroke sensor. Also damaged parts of hydraulic piston motor were found from the vehicles with increasing rotation feedback amount of hydraulic fan system.