Summary
Manufacturing is under pressure. Shorter lead times, greater customer complexity and rising demands for flexibility mean that operational resilience has gone from being a competitive advantage to a condition of survival.
This analysis identifies the five most widespread operational challenges in manufacturing, maps the trends shaping 2026, and points to the factors that research and field experience consistently link with lasting improvement. The analysis is based on operations-management literature and recognised research — not on surveys or our own client data.
Introduction
Most manufacturers know that something isn’t working. They repeat the same problems. Operations depend on individuals. Improvement projects deliver short-lived results and disappear again. A condition well known in business and well documented in research.
W. Edwards Deming argued as early as 1986 that over 94% of all operational problems stem from systems and structures — not individuals. Nelson Repenning and John Sterman (MIT) demonstrated in 2001 how organisations under pressure fall back into old patterns, even after successful improvement programmes — what they called the “capability trap.” John Shook described in Harvard Business Review (2010) how behaviour — not attitudes or knowledge — drives cultural change.
These insights are not new. But they are still not broadly implemented. This analysis attempts to answer why — and what can be done about it.
Part 1: The five biggest operational challenges
Based on operations-management literature and field observations across manufacturing, trades and services.
Invisible capacity loss
The most systematic finding in manufacturing OEE research is that companies underestimate their actual losses. Studies by Jonsson & Lesshammar (1999) and Nakajima (1988) show that average OEE in European discrete manufacturing lies between 40-60%, while the organisations themselves typically estimate 70-80%. The gap is not down to deliberate deception — it is because the losses are hidden.
Micro-stops (under 5 minutes), waiting time at material changes, informal changeover time and speed-reduction losses are rarely recorded systematically. The result is that you optimise what you can see and let the big losses remain invisible. Value stream mapping (VSM), as popularised by Rother & Shook (1998), consistently shows that the actual lead time for a given process is typically 5-15 times the pure processing time.
“The purpose of learning to see is to find and eliminate waste, not to create perfect-looking maps.”
— Mike Rother & John Shook, Learning to See (1998)
The solution starts with precise measurement — not estimation — and requires the organisation to be willing to document the uncomfortable picture that emerges.
Improvements that don’t anchor
Perhaps the most documented challenge in the Lean literature is that improvements don’t hold. Repenning and Sterman (2001) call it the “capability trap” — under pressure, organisations revert to old work patterns, even when they know better. The phenomenon is not a question of motivation or knowledge; it is a systemic phenomenon driven by short-term time pressure.
Hiroshi Osada (1991) and Masaaki Imai (1986) both pointed out that standardisation is the primary stabilising mechanism in a Lean system. Ohno put it as: “Without standards, there can be no improvement.” Yet standard work is one of the least implemented elements in Western Lean roll-outs.
The research points to three stabilising mechanisms: (1) management routines that hold on to the new behaviour, (2) visible measurements that make relapse obvious, and (3) structured follow-up that intervenes when the process drifts.
Frontline management as the bottleneck
The production leader — the team leader, the supervisor, the shift supervisor — is the most critical and most underrated management role in manufacturing. Research by Womack, Jones & Roos (1990) from the MIT study of the car industry showed that effective frontline managers were the single strongest factor explaining productivity differences between comparable plants.
Yet frontline managers are typically chosen for technical competence, not leadership ability, and are rarely given formal management training. They are expected to “learn it themselves.” Mike Rother describes in Toyota Kata (2010) the systematic coaching pattern Toyota uses to develop frontline managers — a pattern almost never copied in the Western implementation of Lean.
“The way to change culture is not to first change how people think, but instead to start by changing how people behave — what they do.”
— John Shook, Harvard Business Review (2010)
Digitalisation without an operational foundation
MES systems, IoT sensors and production ERP are increasingly available to mid-sized manufacturers. But research and practice consistently show that digital systems amplify what already exists — and that works both ways.
An organisation without stable processes and clear lines of responsibility gets more complexity, more data and more noise — not better operations. Pascal Dennis (2002) describes this as “automating a mess.” Conversely, an organisation with stable processes, clear standards and working escalation can use digital tools to create real leverage: earlier warning signals, better resource utilisation and faster decisions.
The conclusion is not that digitalisation is wrong. It is that the sequence is decisive: stable operations always came before digital operations.
Person-dependency and concentration of competence
Many manufacturers have critical knowledge concentrated in individuals — typically those with long tenure and broad experience. This model works as long as those people are present. It breaks down at absence, resignation or retirement.
The Skill Matrix approach (Shingo, 1989; Monden, 1993) is the systematic answer: map competences visually, identify vulnerabilities, and design targeted upskilling focused on broader competence coverage. It is not about making everyone the same — but about ensuring that critical processes don’t depend on one person’s presence.
Part 2: Key trends in 2026
Trends shaping the operational agenda for manufacturing.
Lean and the digital twin converge
Digital twin technology — virtual models of physical production processes — makes it possible to simulate improvements before they are implemented. For companies with stable standard work and precise process data, this is a powerful lever. For companies without that foundation, it is yet another unused system.
Energy efficiency as an operational target
Energy costs are increasingly a competitive parameter in manufacturing. Organisations that already work systematically with OEE and loss analysis are best positioned to identify energy-related sources of waste — because energy losses and process losses typically correlate.
Retention through better work systems
A shortage of skilled labour pushes companies to make the working environment attractive. Well-defined processes, clear allocation of responsibility and visible management follow-up are documented to be linked with lower staff turnover (Hackman & Oldham, 1976 — job characteristics model).
Shorter improvement cycles
Toyota Kata's improvement cycle (Rother, 2010) — weekly experiments towards defined target conditions — is gaining ground as an alternative to the large, drawn-out improvement projects. Shorter cycles give faster learning and reduce the risk of losing momentum.
Part 3: What separates those who succeed
Three factors consistently linked with lasting operational improvement — across industries and company sizes.
1. Management behaviour — not intentions
John Shook (2010) and Mike Rother (2010) agree: it is what managers do that shapes the behaviour in the organisation — not what they say they mean, or what culture they want. Managers who respond consistently to data, who go to the root of problems rather than settling for quick fixes, and who keep fixed follow-up rhythms build an improvement culture from the ground up.
Deming (1986) called it “constancy of purpose” — a stable direction that holds firm, even under short-term pressure. It is rare, because organisational systems typically reward short-term firefighting, not long-term system improvement.
2. Standard work as the stabiliser
Ohno and Shingo both identified standard work — documentation of the best known way to perform a given task — as the foundation beneath all improvements. You can’t improve what varies randomly. Standardisation is not bureaucracy; it is the foundation that makes improvement possible and measurable.
Imai (1986) argued that kaizen (continuous improvement) presupposes standardisation: “Standardisation is necessary to maintain the gains achieved through improvement.” Companies that have stable standards see improvements that hold. Companies that don’t have them see improvements that slide back.
3. Rhythm and escalation as governing infrastructure
Daily Management structures (tier meetings, board control, escalation matrix) give the organisation a control system that catches problems before they grow large. The Lean Enterprise Institute and Shook (2008) describe, in “Managing to Learn,” the role of A3 thinking in creating a culture where problems are raised — not hidden.
What matters is not the systems in themselves — it is the rhythm they create. Organisations with fixed weekly and daily follow-up rhythms respond systematically to signals and build trust from the top down. Without rhythm, even the best systems are non-functioning.
Part 4: Recommendations
Five concrete actions with a high likelihood of lasting effect — based on research and field experience.
Map the actual current state
Not what you think happens — what actually happens. VSM analysis, OEE measurements and time/motion studies give the precise picture. Prioritise the three biggest loss points. Everything else is second order.
Establish standard work for the ten most important processes
Identify the processes with the greatest variation and the greatest consequence. Write standards short enough to be followed. Test them. Revise them when better methods are found. Standards are living documents — not control.
Introduce a simple Daily Management structure
A daily 10-minute tier 1 meeting at the board, with four questions: What happened yesterday? What are the deviations? What is the plan today? What needs to be escalated? That is enough to begin.
Invest in frontline management
Give your team and shift leaders concrete management tools, not just technical knowledge. The Toyota Kata coaching cycle is a documented example. Alternatively: define what good management behaviour looks like, measure it, and give systematic feedback.
Stabilise before you digitalise
Map which processes are stable enough for digital systems to add real leverage. For the rest: stabilise the process foundation first. An MES system on unstable processes is a more expensive version of the same problems.
Sources
- Deming, W.E. (1986). Out of the Crisis. MIT Press.
- Dennis, P. (2002). Lean Production Simplified. Productivity Press.
- Hackman, J.R. & Oldham, G.R. (1976). Motivation through the design of work. Organizational Behavior and Human Performance, 16(2).
- Imai, M. (1986). Kaizen: The Key to Japan's Competitive Success. McGraw-Hill.
- Jonsson, P. & Lesshammar, M. (1999). Evaluation and improvement of manufacturing performance. International Journal of Operations & Production Management, 19(1).
- Monden, Y. (1993). Toyota Production System: An Integrated Approach to Just-In-Time. Industrial Engineering and Management Press.
- Nakajima, S. (1988). Introduction to TPM. Productivity Press.
- Repenning, N. & Sterman, J. (2001). Nobody ever gets credit for fixing problems that never happened. California Management Review, 43(4).
- Rother, M. (2010). Toyota Kata. McGraw-Hill.
- Rother, M. & Shook, J. (1998). Learning to See. Lean Enterprise Institute.
- Shingo, S. (1989). A Study of the Toyota Production System. Productivity Press.
- Shook, J. (2008). Managing to Learn. Lean Enterprise Institute.
- Shook, J. (2010). How to Change a Culture: Lessons from NUMMI. MIT Sloan Management Review, 51(2).
- Womack, J., Jones, D. & Roos, D. (1990). The Machine That Changed the World. Rawson Associates.
This analysis was prepared by Prozanta and is based on publicly available academic material. It does not constitute legal or financial advice.