Seeq Is Changing the Paradigm of Analytics to Empower Engineers
There is a current industry focus on "prescriptive analytics,” the ability to identify assets that will soon need maintaining and to prescribe the best actions to take. It sounds like the holy grail, doesn’t it? But it’s not.
Prescriptive analytics is simply the wrong paradigm because there is no way an application could be preloaded with every possible piece of context to inform the right decision in the moment. All the relevant context – pricing, resources, schedules, objectives, and more – is too broad a consideration set.
The better approach is to empower engineers with access to contextual data and advanced analytics so in the moment they can decide what to do next within the big picture of what will drive the desired production or business outcome.
This approach puts corporate objectives – profitability, for example – squarely front and center when critical decisions need to be made. This is a key distinction. Let’s take a closer look at prescriptive analytics and why it should not be held up as the ideal.
Understanding the age of analytics:
Big data/analytics has unquestionably changed the industrial world (along with everything else). In recent years, the goal of analytics has expanded from descriptive to diagnostic, predictive, and now prescriptive analytics.
- Descriptive analytics are by definition backward-looking; they describe what happened. The vast majority of businesses across industries have used and still use descriptive analytics in the form of production dashboards, reports, and KPIs.
- Diagnostic analytics seek to identify why something happened. The benefit is to understand what happened to avoid downtime and affiliated losses in the future. Correlation, causation, and regression are popular approaches.
- Predictive analytics enable engineers to identify what will likely happen, based on current-state data, enabling corrective action to be taken in time to either change, avoid, or plan for the outcome.
- Prescriptive analytics attempt to define what should happen given a set of inputs. The challenge is it is impossible to anticipate every eventuality and the impact of different decisions on business and production priorities.
Everything in context
When predictive analytics are combined with contextual data and powered by an advanced analytics application like Seeq, engineers can make an informed decision on what to do in any situation based on high order priorities. For example, the price of our product is X, the safety scenario is Y, the tradeoff vs. operating at an impaired capacity is Z. With access to the right contextual data and advanced analytics, engineer can incorporate the different data sets along with business objectives to arrive at the best outcome in that moment. This is much more powerful than starting with canned prescriptions that don’t – and can’t – include all relevant information.
This approach – call it profitability analytics – is an important improvement over the way things are typically done. We have seen manufacturers throw away batches of drugs that are worth hundreds of thousands of dollars, for example, because it would take less time to restart the batch than it would to do the required analytics to improve the batch outcome. It was simply too time consuming and expensive to do the analytics. It got to the point where it was cheaper not to use the heavy-weight analytics tools – they just took too long to yield insights.
Process engineers need an application that is accessible to them to accelerate what they are trying to do. At Seeq, our goal is to empower and enable engineers to do, not change the critical tasks they have always done. In the moment, profitability analytics can lead to optimized outcomes, which is a more realistic and beneficial outcome for organizations.