Transform your Data with Sabisu Analysis

We are pleased to present some exciting new changes to Jidoka in the form of Chita, Sabisu’s analysis ‘sandbox’. Several new charts have been added making it easier to analyse your data and share that analysis with others.

Chita can take data directly from Data Historians such as Aspen Tech IP21 and OSI Soft PI enabling you to view and interrogate it as you never have before.  Chita can show you patterns, trends and anomalies in the data you may not previously have been aware of. Not only can it show you current trends; it can predict future trends which you can use to plan and optimise future production or analyse past production.

If you’d like to get Chita or would like further insight into how it can revolutionise your data please contact us.

Introducing Dynamic Time Warping

Dynamic Time Warping (DTW) provides a method of aligning two time series in an irregular way.

The time series can be stretched in some parts and compressed in others to achieve the best match between the two. The chart background shading shows when either curve progressed more quickly than the other, or both curves progressed at the same rate.

time warp

Welcome Curve Alignment

Curve Alignment or Linear Transform as it is also known is the most common method of aligning two time series. One curve is transformed to fit the other as best as possible, with the Shifts and Scales used in this transformation describing how this has been done. In Sabisu the second curve is transformed to fit the first.  In the chart below TDPM:2652 has been scaled to 16% of its original size and shifted to 12 hours later than it originally was.

Curve allignment

For example, if you produced 100 tonnes of product last May but this May you only produced 50 tonnes, Curve Alignment could show that production shifted and full production started later (or earlier) or if the data is stretched it could show that you were producing at half the rate of last year.

Statistics Added to Basic Data Chart

The Basic Data Chart gives you a “normal” view of a tag from a data historian, for further analysis you can use the other Sabisu Charts to get a more in-depth look at your data.

data

Statistics have been added into the sidebar for the basic data view. These show details such as the current deviation, the change of value over the visible period, the average and the standard deviation. Help text is also available in the side bar for more complex algorithms, with descriptions of the analysis that has been performed on a particular chart. You can also select a range of data on any chart and zoom in to get a better look at what occurred.

analysis

This analysis capability is available for all data; project and operational.

With these changes we hope Chita will help you analyse your data more effectively to improve production rates, quality, identify best practice, reduce costs or control risks.

Contact us

We’re always interested in hearing from you with any comments or suggestions, feel free to get in touch.

Start typing and press Enter to search