Caterpillar announced Cat® Asset Intelligence 5.0, a new version of its condition monitoring system, at OSISoft’s PI World in Barcelona.
The asset intelligence (AI) system helps users predict and avoid failures, optimise maintenance, reduce fuel and operational expenditure, ensure compliance and create transparency, Caterpillar said. AI is particularly suitable because it is built around six pillars that make it compliant with ISO 13374’s standards for condition monitoring: data acquisition, data manipulation, state detection, health assessment, prognostic assessment and advisory generation.
Cat AI 5.0 will improve on the current version, 4.8, in two main ways, Caterpillar Marine business development manager David Shannon explained to Marine Propulsion on the sidelines of the event. Firstly, it allows customers to make their own edits to information screens advising courses of action based on the data being processed, and secondly, users can make edits to the model itself.
While Caterpillar develop the core model based on analytics, users will now be able to tweak these models to better suit their particular applications. Mr Shannon gave the example of a user whose vessels operate in ambient water temperatures warmer than those used to create the model. This could result in false positive warnings, but with AI 5.0 the user can amend the model to factor this in.
“Our two criteria of being scalable and flexible are being met,” said Mr Shannon – scalable because changes to models can be easily rolled out fleetwide, and flexible because users can make changes to models based on their own applications of assets.
A big focus in AI 5.0 is moving from the concept of ‘how it works’ to ‘what it does’, said Mr Shannon, explaining that there is a history in condition monitoring of concentrating on how modelling functions work rather than actually explaining to users what capabilities it can bring. AI 5.0 will attempt to change this with a superior user interface.
Another noteworthy feature of Cat AI is that it is agnostic, meaning it is compatible with equipment from other OEMs and industries outside of marine and not just Caterpillar products.
Caterpillar’s analytics and modelling capabilities benefit from the company’s involvement with the US Navy, Mr Shannon said. From the late 1990s through to 2010, the company was responsible for close to 4,000 discrete assets for the US Navy and has more than 40 m hours of data and analytics as a result.
Using a suite of OSISoft’s PI products including servers and connectors, condition monitoring can be presented to shipboard users such as chief engineers and crew using Caterpillar’s onboard viewer, as well as being transmitted to head office for remote monitoring and analytics to aid in fleet advisory and customer management, for example. It was also announced at the conference that Caterpillar would be integrating OSISoft’s Connected Services into the CAT AI platform.
Caterpillar’s AI 5.0 is targeted to hit the market at the end of Q1 2019.