Designing for Humans: Viewing DFM and Industrialization Through the Lens of the Fitts MABA–MABA List

• Hardware Development
• Product Design
• Optimization

Low manufacturing yield, recurring quality escapes, and other stubborn production issues are usually viewed through the lenses of immature processes, inadequate operator training, and supplier variability. However, even in scenarios where these factors are contributors to the issues, they usually do not tell the whole story.  A fundamental factor that is always missing in these conversations is the lack of manufacturing-focussed human-factors considerations in product design and design for manufacturing.

                                                            Fig1: The Red Bead Experiment

One of my favorite quotes, recently, is by W. Edwards Deming from the Book, “Out of the Crisis”  - 

“The supposition is prevalent the world over that there would be no problems in production or service if only our production workers would do their jobs in the way that they were taught. Pleasant dreams. The workers are handicapped by the system, and the system belongs to the management.”

For product manufacturing, I like rephrasing the ownership boundary bit of the quote as; “the assemblers are handicapped by the product’s design, which of course belongs to the design engineers”.

Manufacturing is, largely, still a human-centered activity with human assemblers, technicians, inspectors, and operators, yet many products are designed with a silent expectation of machine-level invariability from the human elements in the production process, without proper consideration for their human limitations. 

In this article, I will be highlighting how a meaningful portion of manufacturing and industrialization pain is rooted not in execution failure, but in design shortcomings and how reviewing the design through the lens of the Fitts MABA-MABA list as a guard rail during DFM, can help identify and remove latent failure modes long before they show up as yield loss or rework.

Manufacturing as a human system

The systems that shape day‑to‑day work on the manufacturing floor are largely defined by the engineers who designed the product being manufactured through product architecture, part geometry, tolerance choices, and through the structure of DFM decisions, MEIs, SOPs, etcetera. When those systems fail to account for human strengths and limitations, capabilities, and shortcomings, assemblers and operators are left to compensate, most of the time, unsuccessfully.

As evidenced by design decisions, design engineers often subconsciously assume idealized assemblers and manufacturing conditions expecting:  

• Perfect attention, indefinitely
• Zero Drift: No degradation due to fatigue, stress, or time pressure
• Perfect interpretation of written instructions
• No cognitive overload from multi-variant builds
• Uniform skill levels across operators and sites

When these assumptions fail, the reaction is typically to add inspections, increase training, or tighten supervision. While these actions are sometimes necessary, they treat symptoms rather than causes.

"Inspection to improve quality is too late, ineffective, costly. Quality comes not from inspection, but from improvement of the production process". - W. Edwards Deming 

A useful, proactive, upstream question that we rarely ask is: Does this design decision lead to manufacturing tasks that violate basic human strengths and limitations? Answering that requires a way to reason about human performance without turning manufacturing into an academic exercise. This is where the Fitts MABA–MABA list becomes useful.

A Quick Refresher: The Fitts MABA–MABA Principle

Developed by Paul Fitts, the MABA–MABA ("Men Are Better At / Machines Are Better At") list is a foundational principle in human factors engineering. It emphasizes the importance of evaluating tasks and assigning them to man or machine based on what they excel at. 

According to the list, humans generally perform well when tasks involve:

• Judgement under uncertainty
• Visual and pattern‑based recognition
• Adapting to variation
• Handling exceptions and edge cases

Machines, on the other hand, excel when tasks require:

• Sustained repetition
• Quantitative consistency
• Precise force, timing, or alignment
• Long‑term vigilance without drift

While the capability of machines has improved significantly since the list was originally proposed, its evaluation of human capabilities remains valid. It provides a good starting point and a disciplined approach to evaluating the human factor impacts of a product’s design on those in charge of the production. 

DFM Beyond Geometry and Tolerances

Traditional DFM focuses heavily on physical feasibility: part count, access, tolerances, tooling constraints, and cost. While these considerations are necessary, they are not sufficient. A human‑centered extension of DFM asks additional questions like:

• Is correctness encoded physically or procedurally? (Does it snap together only one way, or do I need a 2-page PDF to explain the orientation?) Poka-Yoke as a manufacturing gospel is so widely preached yet sparingly practiced. 
• Does this step rely on human precision, or judgement? (Am I asking an operator to apply exactly 2.5Nm of torque by "feel," should we be providing a calibrated tool? Can the design handle the invariability? will damage caused by additional 0.5Nm of torque be visible?)
• Is the "Signal-to-Noise" ratio high? (Can the inspector easily see the defect, even after an 8-hour shift, or is it a "subtle visual distinction" buried under a heat sink?)

This reframing and additional questions prevents the designer, from making a decision that relies on an operator’s 20/20 vision, their memory of a SKU list, or their tactile sensitivity at 3:00 PM on a Friday.

Practical Manufacturing Implications and Examples

human‑factor‑driven issues in production tend to appear when:

• Correctness depends on subtle visual distinctions
• Tasks require continuous precision over time
• Instructions rely heavily on interpretation rather than recognition
• Operators must remember parameterized differences across similar steps

To help drive these points home some example scenarios are evaluated below; 

1.  Proprioceptive Fatigue in Blind Mating:

• The Issue: Design uses a "blind mate" internal connection (like a battery lead or a mezzanine connector) where the operator cannot see the mating interface.
• Factory Floor Reality: While "feel" is a human strength, it degrades rapidly with fatigue. After 200 units, the operator’s proprioception (the sense of self-movement and body position) dulls. They may miss a "crunchy" mate or a partially seated terminal, leading to latent "intermittent" failures that are a nightmare for RMA teams.
• The MABA Fix: If a connector must be blind-mated, it should at least have lead-in chamfers that are 3x the maximum expected misalignment and a locking "click" that is audible above the ambient 70dB factory floor noise.

2. Low Visual Signal-to-Noise Ratio (SNR) in Inspection

• The Issue: Design has critical inspection points (like a gasket seal or a locking tab) where the "correct" state is visually subtle or hidden behind more prominent components.
• Factory Floor Reality: An inspector’s ability to catch a defect is a function of the Visual SNR. If you are looking for a black O-ring seated in a black plastic groove, buried 40mm deep in a chassis, the "signal" (the defect) is indistinguishable from the "noise" (the shadows and surrounding geometry). Expecting a human to catch a 0.5mm O-ring gap over a full shift is a MABA-MABA violation. The human eye is a change-detector, not a precision scanner.
• The MABA Fix: Design for "High-Contrast Verification." If a connector must be locked, use a bright, "Day-Glo" orange locking tab that is only fully covered when the connector is 100% seated. This transforms a high-effort "Is this seated?" search into a low-effort "Do I see any orange?" pattern recognition task. You are essentially increasing the signal power while dampening the background noise.

In general, applying the MABA–MABA lens typically leads teams toward design choices that:

• Encode correctness physically rather than procedurally
• Reduce the need for real‑time human judgement
• Shift precision demands into tooling or geometry
• Make incorrect states difficult or impossible to achieve

Rethinking Yield Loss and Inspection

High scrap rates are usually accompanied by statements like: “The assemblers are not doing a good job” or “We need to add another layer of inspection.” Viewed through the MABA–MABA lens, recurring errors are rarely a personnel problem but a signal of design issues.

Training and inspection in those scenarios are also just asking people to compensate for a design that demands constant vigilance, perfect memory, or sustained precision, traits humans are simply not optimized for. In many cases, an inspection step exists not because inspection is inherently valuable, but because the design failed to encode correctness directly into the product or process.

Designs should more often than not rely on geometry, fixtures, and constraints that make the correct outcome the natural, and often the only, possible outcome. When a process depends on human vigilance to prevent defects, the defects are already designed in.

Where This Fits in Industrialization

The MABA–MABA perspective is most effective when applied early, while design flexibility still exists:

• In design reviews and FMEAs
• In DFM and industrialization, while drafting MEIs and SOPs
• During FAI and pilot builds, with deliberate observation of operator behavior
• When deciding what to automate and what to leave manual

It integrates naturally at these stages, and introduces a dimension that is often left implicit during them: whether responsibility for correctness has been assigned to the appropriate agent.

Beyond Manufacturing

Although this article focuses on industrialization and manufacturing, the same thinking applies across the product lifecycle:

• Test and calibration workflows
• Service and maintenance procedures
• Field installation and commissioning
• Operator‑facing interfaces

Any system that depends on people will eventually reflect human strengths and limitations, whether designers plan for them or not.

Closing Thoughts

Good DFM is not only about reducing part count or tightening tolerances. It is also about acknowledging that humans are an integral part of the manufacturing system and ensuring the design reflects a level of expectation that matches their limitations and capabilities. 

The Fitts MABA–MABA list provides a practical way to reason about that reality. By aligning manufacturing tasks with human strengths and reserving machine‑like demands for machines, or redesigning them to meet human capability, teams can reduce variability, improve yield, and build processes that work reliably on the factory floor.

Human factors are not an optional layer added after production problems appear. They are a design input, and treating them as such is a hallmark of mature industrialization.

References

1. Fitts, P. M. (Ed.). (1951). Human engineering for an effective air navigation and traffic-control system. Washington, DC: National Research Council.
2. Deming, W. E. (1982). Out of the crisis. Cambridge, MA: Massachusetts Institute of Technology, Center for Advanced Engineering Study.

Sign in