I recently published a sample dashboard created using Wonderware Intelligence software.
The dashboard has been created around a scenario of a manufacturing company described below:
Such a company would have a number of information systems such as:
- financial accounting system (or invoicing system) – such system would provide information about sales volume along with profit margin attained
- production tracking system (implemented as either, off-the-shelf solution such as Wonderware Operations, or as a 3rd party or custom solution) – this system could provide information about planned and actual production volume broken down by product, production line, machine etc. This system could also provide some quality information (e.g. material dimensions or other characteristics)
- traditional HMI/SCADA system – which could provide some information about equipment utilization (e.g. downtime vs. uptime) and energy consumption (so that comparison between nominal energy consumption and the actual usage could be performed)
Coming our of the current recession, the leadership of the company would like to get some information about their system to make smart decisions. They would like to know current and historical values of:
- financials – sales and margin, broken down by customer and product
- production schedule attainment broken down by shift
- equipment efficiency by production line and individual machines
- energy usage by line and machine
This is where using Wonderware Intelligence software we can create data repository that combines and correlates information from all the data sources listed above.
2. Wonderware Intelligence software – server side
Wonderware Intelligence comes with a set of connectors to other products such as:
- Wonderware Historian (formerly called IndustrialSQL Server or InSQL),
- OSI Pi,
- obviously generic Microsoft SQL Server
- and CSV files.
These connectors allow the Intelligence Server to import data on regular basis from the production system and put it in the Intelligence own data store (this way when run the reports we don’t tax the production system and data retrieval is much faster as the data is pre-aggregated).
Information model created in Wonderware Intelligence has a form of multidimensional cube that allows us to slice and dice information according to our criteria. The model has three major components:
- data sources – they define how to connect to our data (e.g. financial accounting system or plant data historian system)
- dimensions – they define our criteria (such as shift, product, customer etc)
- measures – they define value we want to measure or analyze (sales volume, production volume etc)
Shown here is a sample screenshot of Wonderware Intelligence data model in the ArchestrA IDE (Integrated Development Environment) used by Wonderware System Platform.
After all these data sources, dimensions and measures have been defined, users can connect to them using Wonderware Intelligence Analytics Client – very easy to use and very powerful software that can be used directly by people who need the reports.
3. Wonderware Intelligence software – power user experience
In the Intelligence Analytics Client environment, the user is presented a very intuitive data model that consists of dimensions and measures (as shown here). No need to know Transact SQL programming language, no need to get access to the production system, no need to know how to calculate daily total based on 1-minute data etc.
The user is presented with a list of dimensions and measure and all we need to do is drag appropriate parameters to the worksheet. We can add filters, change sorting, change colors and much more, all using point and click user interface.
What’s important is that the report users can create reports for themselves without waiting for their IT departments. They can save the reports locally in a way similar to saving Excel files and they can share them with other users.
The user can create multiple worksheets and then lay them out on a dashboard.
Finally the dashboard can be published to the Intelligence Dashboard Server – just like the dashboard with which we started this posting.
4. Intelligence Dashboard
So let’s talk about the presented dashboard. It consists of four distinct parts:
- Sales and Margin Contribution (upper left chart) – this report shows sales distribution per customer and per product expressed as a square – the bigger the square size is, the greater the sales volume is. On top of that, the color of the square represents our margin – red means low margin, green means good margin. How easy is it to figure out that we’re having a bit of a margin issue with products COO250 and COO378 – for some reason our margin on them is lower than on other products.
- Equipment Efficiency (upper right chart) – this report shows efficiency for each machine in the system. So it looks like Coffee Line has much lower efficiency comparing to any other line.
- Production Schedule Attainment (lower left chart) – this report shows how each shift attains their given production schedule. Well, it looks like the night shift may not be pulling its weight as much as it should.
- Energy Deviation (lower right chart) – this report show the difference between expected energy usage and the actual energy usage.
All these concepts are very similar to multidimensional business intelligence systems such as Microsoft Analysis Services. But few things make Wonderware Intelligence look interesting:
- most configuration steps can be performed without advanced programming knowledge.
- conceptually Wonderware Intelligence’s model is a cube, but it’s implemented using a de-normalized SQL Server database which makes it really easy to query using regular client tools
- finally, the dashboard creation environment is very easy to use by non-IT people (e.g. production engineers, quality personnel etc.)
Performance Information Technologies provides consulting services related to reporting, plant & business intelligence and technology integration solutions for your plant or process. Contact us at firstname.lastname@example.org or (425) 440-0626.