When the Data Says One Thing and Your Operator Says Another with Michael Dorman
What happens when your digital record says one thing, but the experienced operator standing on the plant floor says something else? Who do you trust? As manufacturers invest in AI, automation, digital twins, sensors, and increasingly sophisticated data systems, that question will matter a lot more. Technology can capture an extraordinary amount of information about a manufacturing facility. But data doesn’t automatically replace the experience of the people who know what’s behind the walls, where the undocumented repair was made, why that machine was moved six years ago, or when something simply doesn’t look right. In this episode of The Manufacturers’ Network Podcast , Lisa Ryan talks with Michael Dorman, co-founder and CFO of Primera Works , about what happens at the intersection of manufacturing data, facility knowledge, technology, and human judgment. Michael brings more than 27 years of experience in reality capture, LiDAR, building information modeling (BIM), digital twins, construction documentation, and technology for the built environment. Together, Lisa and Michael explore what technology can preserve, what still lives in the heads of experienced employees, and what manufacturers risk losing when those people walk out the door. The central takeaway: Manufacturing technology works best when accurate facility data and experienced human judgment work together. Digital twins, BIM, LiDAR, AI, and automation can preserve and organize information, but manufacturers still need experienced employees who can interpret that information, challenge it when something doesn’t look right, and add the institutional knowledge that technology alone can’t capture. Manufacturers are investing heavily in AI, automation, sensors, digital twins, and data systems. But more technology doesn’t automatically create better institutional knowledge. If the underlying information is incomplete, outdated, locked inside proprietary systems, or disconnected from what experienced employees know, technology can simply make bad information easier to access. And there’s another problem: employees may be reluctant to share decades of hard-earned knowledge when “knowledge transfer” sounds suspiciously like “teach us everything you know so we can replace you with somebody cheaper.” Lisa and Michael discuss why psychological safety matters in knowledge transfer and how leaders can change that conversation. Instead of treating experienced employees as databases waiting to be downloaded, organizations can position knowledge capture around legacy, expertise, and making the operation stronger for the people who come next. When Experience Challenges the System One of the most important questions in the episode is simple: If the digital record says one thing and the experienced employee standing on the plant floor says another, who wins? Michael’s answer is clear: experience still matters. The goal isn’t to choose between technology and people. It’s to reduce the gap between what the system says and what experienced people know by giving employees relevant, accurate information they can actually use. That becomes increasingly important as manufacturers adopt artificial intelligence, automation, digital twins, BIM, reality capture, and other Industry 4.0 technologies. A sophisticated system isn’t automatically a useful system. Michael points out that manufacturers can invest in technology packed with functionality while employees use only a fraction of it. Lisa brings that down to a much more practical starting question for plant leaders: “What sucks about your job? How can we make it better?” Sometimes the smartest digital transformation strategy starts there. The Knowledge Hidden Inside Your Facility Institutional knowledge isn’t limited to how employees run equipment or make the product. Experienced maintenance technicians, operators, engineers, facilities leaders, and other longtime employees may know: what’s actually behind the walls where repairs and modifications were made which drawings and as-builts are no longer accurate why equipment was moved or changed which workarounds became part of everyday operations what has changed since the facility was originally designed which conditions matter even though they were never formally documented When those employees retire or leave, that knowledge can disappear with them. Michael explains why manufacturers need an ongoing process to document and digitize facility conditions, rather than treating reality capture or BIM as a one-time project. Plants change. Equipment changes. Processes change. Even seemingly minor facility changes can eventually matter to maintenance, safety, construction, risk management, and future capital projects. A Digital Twin Doesn’t Need to Be Rocket Science Digital twin has become one of those phrases that can mean almost anything depending on who’s selling it. Michael offers a much more practical definition: determine which information the owner actually cares about and which needs to move from the physical environment to the digital environment. That’s the digital twin that matters. Manufacturers don’t necessarily need the most sophisticated model available. They need accurate facility information that employees can understand, trust, update, and use. For organizations beginning this process, Michael recommends starting with something far less glamorous than AI: Accurate as-built information. If you don’t have reliable drawings of your current facility, start there before worrying about creating an elaborate 3D digital environment. Human Judgment Still Has a Job Technology can capture enormous amounts of information. It can speed up decisions, organize data, create digital representations of physical environments, and make facility information easier to retrieve. But it doesn’t eliminate the value of experience. That matters when an operator notices something that doesn’t look right, a maintenance technician knows the drawing doesn’t reflect reality, or an experienced employee challenges what the system is telling them. Lisa connects this to another manufacturing leadership problem: the employee who stops a machine because something feels wrong takes a visible risk. If they’re wrong, everyone sees the downtime. If they’re right and prevent a failure? Nothing happens. Nobody walks through the plant announcing, “Hey, at 2:17 this afternoon, nothing happened.” That’s exactly why manufacturers need cultures where employees are willing to speak up, question the data, report near misses, and use their judgment instead of automatically deferring to the system. Facility Data Is Also a Business Risk The conversation also explores the connection between accurate facility information and capital expenditure, construction, safety, and operational risk. When manufacturers build, expand, renovate, or modernize facilities, owners often rely heavily on architects, engineers, contractors, and technology vendors to manage information. Michael argues that manufacturers need to take ownership of the data they expect to retain. After all, when the project ends, the manufacturing company is the one left operating the facility. Clear data specifications, accurate documentation, current as-builts, and usable asset information can help manufacturers protect the long-term value of their capital investments and reduce the risk of having to reconstruct their own facility history years later. Key Questions Answered in This Episode What should manufacturers do when data conflicts with an experienced operator’s judgment? What institutional knowledge is most likely to disappear when experienced manufacturing employees retire? What should manufacturers capture before creating a digital twin? How can manufacturers use AI and automation without losing human judgment? Why are accurate as-built drawings important for manufacturing facilities? How can BIM, LiDAR, and reality capture help preserve institutional knowledge? How can manufacturers encourage experienced employees to share what they know? Why does data quality matter when manufacturers invest in AI and automation? How can better facility information reduce risk during plant expansions and capital projects? In This Episode, You’ll Learn Why retiring employees can take critical facility knowledge with them How manufacturers can improve knowledge transfer without making experienced employees fear they’re documenting themselves out of a job Why psychological safety matters when capturing institutional and tribal knowledge What BIM, LiDAR, reality capture, and digital twins can preserve Where human judgment and experience still matter Why accurate data matters more than simply collecting more data How outdated as-builts create problems for maintenance, construction, safety, and capital projects Why manufacturers should own and specify the facility data they expect to retain How overly sophisticated technology can create more friction instead of more value Why plant leaders should start with the information employees actually need How manufacturers can begin documenting their physical facilities before experienced employees retire Why AI and automation work better when human expertise remains part of the decision-making process About Michael Dorman Michael
