TrendMiner - Software for Predictive Analytics for the Process Industry

Prevent

Predicting process behavior with TrendMiner does not require a data scientist. Your process and asset engineers can immediately apply their subject matter expertise to solve potential production issues before they occur or plan preventive maintenance based on process evolution.

Predict

Through the “contextualization of asset performance with process data” early warnings can be extended to the level of predictive maintenance. With the captured knowledge, the organization can receive appropriate instructions what needs to be done or when to schedule the work.

Our next generation predictive analytics give you a glimpse of the future.

TrendMiner enables early warning detection of abnormal and undesirable process events by comparing saved historical patterns with live process data. The software calculates possible trajectories of the process and predicts process variables and behavior before they happen.

Users are shown how far the process has evolved already, plus how TrendMiner predicts the process it will continue to evolve in future. There is no time delay – the results are shown instantly. This gives operators the ability to see if recent process changes match the expected process behavior, and to pro-actively adjust settings when it does not.

Predicting process behavior with TrendMiner does not require a data scientist. TrendMiner is a self-service solution that it designed to be used by subject matter experts. When your engineers can access predictive analytics insights directly, they are able to immediately apply their subject matter expertise to solve potential production issues before they occur.

Using defined fingerprints for best performance, automatic alerts can be sent to the appropriate stakeholders, such as control room personnel or field engineers, to take appropriate action in the event of process deviation.

People can either be warned through notifications or more actively through sending emails with a pre-set message that instructs them what to do given the situation predicted. In this way, the process can be proactively adjusted to avoid waste and ensure product quality.

TrendMiner’s early warnings capabilities can be used for process control alarms to assist operators as they monitor and control processes, and alert them to abnormal situations.

TrendMiner predictive analytics help operators by informing them before the event occurs. The operators can also respond to each situation more quickly using the knowledge that has been captured in TrendMiner. This helps them to control production levels and quality, avoid costly shutdowns and reduce safety risks in a timely manner.

Want to predict key measurements or lab analyses that are expensive to obtain or come in with too much delay? TrendMiner supports creating and deploying soft sensors using an interactive and step by step approach with access to all process data. Help your process experts to build better predictors.

Our approach is based on extensive experience with soft sensor design. It eliminates the need for a data scientist in the loop and reduces the lengthy process of modeling and deployment.

Overall Equipment Effectiveness (OEE) depends on the process in which the asset operates. With TrendMiner, all tags related to the asset and the process in which it functions in can be used to analyze its performance. We call this the “contextualization of asset performance with process data”.

By contextualizing asset performance with process data in this way, data-based predictive maintenance becomes possible. A common example is the fouling of heat exchangers. The process performance directly impacts the fouling and with use of TrendMiner you can predict when cleaning is needed. This can then be scheduled according to production planning to prevent reduced performance.