With growth of digitalization in manufacturing, what does the future hold for Layered Process Audits (LPA)? LPAs today generally verify human conformance to key processes with and outside set of eyes. But with automation, what will humans be doing?... and eyeballs won’t be as accurate as sensors.
Based on a presentation given by Murray Sittsamer and Richard Nave of The Luminous Group at our 2019 Beacon User Summit, this post will discuss the what layered process audit programs might look like in the future with advancement in technology.
Industry 4.0 will be a $155 billion market by 2024, part of a rapid growth in technologies like machine learning and Internet of Things (IoT) revolutionizing the manufacturing industry.
As Industry 4.0 picks up steam in automotive and aerospace manufacturing, we are witnessing an evolution towards more sensors, more data communication and more analytics. And while layered process audits (LPAs) could drive even more powerful quality insights and actions, companies also need to be aware of—and hopefully avoid—the potential downsides.
LPA Questions Will Be Automatically Generated from FMEAs
One area where Industry 4.0 will likely affect LPAs is with auto-generated checklist questions, which could lead to big improvements. Often, we see questions that aren’t meaningful because they aren’t tied to risk and they’re not updated when quality issues occur.
In the future, FMEAs will be able to suggest potential problems to link to LPA questions, pulling from areas like:
- High severity failure modes
- Prevention or detection controls that depend on people
- Prevention controls that involve standards or specs, with controls transmitting specification data directly to eliminate guesswork
There are, of course, potential downsides. For starters, we could see the same problems we see with manual LPAs in terms of poorly structured or poorly worded questions. When A/I protocols write the questions, a certain amount of subtlety is lost, creating concerns about the quality of questions. Teams will also want to maintain expertise in LPAs so they don’t lose those valuable skills—similar to the risks involved with outsourcing too many operations.
Download our free LPA starter checklist
LPA Questions Will Be Automatically Generated from Problem-Solving Software
The second place LPAs are headed in the future is towards automatic generation of questions from a database of resolved problems (e.g. customer issues and engineered solutions). Many quality management system (QMS) solutions currently enable structured problem-solving using tools like 8D, and it’s critical organizations use the resultant solutions and improved controls to update FMEAs and LPAs.
But rather than rely on human memory to get it done, why not automatically generate LPA questions to verify the solution is held in place? Obviously this approach isn’t something you’d use for every issue, but it would be something companies would want to do for problems:
- With a high severity
- That have impacted the end customer
- Where solutions are people dependent or involve a standard or spec
The downside here is similar to the previous point, where questions might be poorly worded or not targeted to the most important aspect of the solution. LPA questions are written once, but different levels of management will use them over and over again. That’s why it’s so critical to take the time to review and ensure the LPA questions are meaningful and easily understood.
LPA Checklists Will Become Production Readiness Indicators
It’s possible LPAs of the future will be a machine-driven set of self-checks acting as production readiness indicators, similar to a pre-flight checklist. Integrated production sensors would be able to check specific items and either give the green light if they sense conforming conditions (e.g.: temperature settings, back-pressures, distance measures, etc., or require immediate corrective action to achieve conformance.
Then there’s the question of whether inputs need to be in perfect conformance to move forward with production, which is a judgment call. Although LPAs can provide an indication of readiness, they still only consist of a few questions—not every single element in an assembly line.
LPAs Will Become “People-less”
Tying into the earlier points, our final prediction is that LPAs may not even involve people in the future. Instead, machines and sensors will undergo automatic checks for readiness status. Those checks will in turn be integrated with the production line, reducing the involvement of actual people in audits.
The big downside here is that eliminating people from audits may negatively impact the quality culture benefits that LPAs provide, including:
- Improved communication between management and operators
- Demonstrating to employees that leadership takes quality seriously
- Opportunities to provide input—and having those suggestions acted upon by management
Truly people-free LPAs would mean losing those relationships and benefits, a big loss that might not show up in a scorecard but could impact long-term performance. That said, it’s also possible that the evolution towards mobile LPAs could give teams more time to focus on interaction with the people who work in the organization, instead of the administrative details of audits and reporting.
No matter what the future holds, those organizations with a strong quality culture will always stand out from the competition. As more and more elements of LPAs become automated, it’s critical that manufacturers find ways to leverage the technology without losing focus on the people it’s meant to serve.
Join the LPA Discussion Group on LinkedIn:
Free discussion board regarding implementing and improving Layered Process Audits and Verifications https://www.linkedin.com/groups/12166916/
Upcoming public training at the Automotive Industry Action Group (AIAG) led by The Luminous Group:
Implementing Layered Process Audits (full-day, September 23, 2019)
Leadership Role in Successful Layered Process Audits (half-day September 24, 2019)
The Luminous Group also provides on-site training and consulting in LPA and other quality topics: