Matthew Littlefield, President & Principal Analyst, LNS Research
There is a lot of hype when it comes to Industrial Internet of Things (IIoT) and Big Data Analytics. Many executives today see the mainstream media marketing campaigns being run by companies like IBM with Watson winning Jeopardy and are instantly interested in seeing how such a tool might be able to solve their business problems.
Although there is undisputed business value in using Big Data Analytics to enable predictive and prescriptive analytics, LNS Research sees most industrial companies are just at the earliest stages of adopting these technologies as part of IIoT pilot projects. For most companies, there is still much work to be done in gaining insights from the data already locked up in their enterprise systems.
In this post we will examine how quality executives can start getting value from the data they already have with an eye to the future, making sure the investments made today will pay dividends in a fully mature IIoT world.
Defining Your Quality Analytics Initiative
Quality executives intuitively understand the questions they want answered to help them do their jobs better:
- How many open Corrective and Preventive Actions (CAPA) are there across my business? How long is it taking to close them? What are my riskiest CAPAs?
- Which suppliers are most disruptive to my manufacturing operations? Which suppliers are driving customer complaints? Which suppliers pose the most risk to my organization?
- Which customers are most and least happy with the quality of our products?
- Is my quality engineering organization accurately predicting product quality and reliability?
- Which nonconformances should I address first and how can I leverage this information to continuously improve?
- How can I leverage my equipment performance data to predict future problems?
However, just knowing the important questions is not enough to have a successful analytics initiative. Quality executives need to document the open questions that exist today as well as leave open the possibility of additional questions in the future. Then, quality executives can begin to map these questions both to the higher-level business goals that are driving the organization as well as to a broader analytics framework that can act as a guide for categorizing the different metrics and tools that will be needed as part of a successful analytics program.
Incorporating Quality Analytics into a Big Data Framework
To map these above questions to metrics and high-level business goals, LNS Research recommends using a model of Operational Excellence to align leadership, technology and quality processes with other enterprise processes. For mapping these questions to a specific analytics framework, LNS Research recommends using a generic framework like Descriptive, Diagnostic, Predictive, and Prescriptive.
With a simple and generic analytics framework agreed upon (and hopefully broadly across the enterprise, not just in quality) it becomes much easier to understand what data sources, data models, analytical techniques, and technologies are needed to answer pressing quality management questions.
It also becomes much easier to understand where to start and what to leave for later as maturity in Big Data Analytics and the IIoT increases. For most companies, the starting point will be in Descriptive and Diagnostic analytics. Quality managers will begin by answering questions like what happened and why using traditional data analysis tools and existing data sources that are included in EQMS, ERP, and PLM.
Avoiding Vendor Pitfalls
In any market as hot as the Big Data Analytics market is today, technology buyers have to be wary of any technology vendor with a new shiny tool looking for a problem to solve. There are many Big Data companies that have been born out of social media monitoring and/or IT equipment monitoring that are now looking for industrial use cases; don’t be caught off guard and become the guinea pig for a technology company that has no industry experience. Also don’t become dazzled by the bright lights; many technology vendors have very slick dashboards but limited experience and out-of-the-box capabilities to get data out and into a robust data model.
In the short term, as quality executives are starting down the analytics journey, most will be best served by choosing a technology partner that understands the existing quality management space and is also forward-looking in the use of Big Data Analytics tools.
On-Demand Webinar: Getting Started with Quality Analytics
Presented by LNS Research
In this webinar, LNS Research dives deeper into their quality analytics framework, presents real world examples, provides survey data documenting best practices for using quality management analytics, and makes specific recommendations for success.