Don't Scale the Old Factory

01

01 — The scaling trap

Growth creates urgency. Urgency makes multiplication attractive.

More orders appear to require more machines. More machines appear to require more operators. More production appears to require more Material Handlers, supervisors, inspectors, maintenance technicians and recruiters. A layout built for today's volume gets stretched, copied or expanded.

That may increase capacity. It can also scale the constraint.

The operating architecture that created today's success can become the ceiling on tomorrow's performance. Compass has made this argument more broadly in its work on enterprise architecture and the next curve: architecture determines performance, and the next stage has to be designed rather than simply added to the current system.

Manufacturing makes that principle tangible. Before multiplying today's work, leaders should determine whether tomorrow's operation should contain that work at all.

That may increase capacity. It can also scale the constraint.

02

02 — Ask a different question

A Model Factory begins with Strategic Intent and a clean-sheet question:

If we were designing this operation for the next growth curve—not for today's installed base—how would material, information, equipment and people flow?

Front-to-Back 80/20™ makes the question economically disciplined. It traces value from customer and product through strategy, resources, organization, process, projects, KPIs, meetings and actions. It asks which small number of products, product families, customers and value streams create most of the value and profitability, and which complexity is consuming flow without creating equivalent value.

The sequence is:

Strategic Intent → required customer outcome → demand, mix and economics → Critical Few product or value-stream pattern → clean-sheet process → material and information flow → automation choices → future organization and roles → capability and management system.

Clean-sheet does not mean ignoring constraints. It means making constraints explicit rather than allowing the current layout, job titles, systems and habits to define the future by default.

That changes the conversation.

Instead of asking how many additional forklift drivers are needed, ask why the material must be moved manually and what the future material system requires.

Instead of asking how many operators are needed for each machine, ask what should be automated, combined, monitored or managed by exception.

Instead of adding inspectors at the end of the process, ask how quality can be designed into the process and detected at the source.

Instead of assuming every current job scales with volume, ask what future work remains after simplification, standardization, automation and digital orchestration—and what new work is created by the technology.

The result is not automation for automation's sake. It is a different operating model built around the work the business actually needs to perform.

03

03 — A Wisconsin proof point: Kohler's long automation journey

Wisconsin provides a useful historical reference. A November 1982 Wisconsin Architect feature described Kohler's new Distribution Center as massive, practical and effective in the storage, handling and shipping of product. The article also documented the unusual scale and logistics required to create the facility. [1]

Nearly four decades later, the University of Wisconsin E-Business Consortium organized a session at Kohler around “Kohler's Automation Journey.” The agenda included a warehouse tour and discussion of warehouse-management systems, inventory control, pick modules and wearable technology. [2]

The lesson is not that a distribution center designed in 1982 contains today's technology. The lesson is more important: state of the art is not a one-time installation. It is an operating philosophy of repeatedly redesigning flow as technology, demand and capability change.

Manufacturing leaders facing their next growth curve can apply the same thought process to the factory.

04

A practitioner’s lens: what 17 Model Factories taught me

I had the opportunity to serve on a three-person steering team responsible for implementing 17 Model Factories around the globe. It was an unusually rich learning experience. It taught me that the decisive work happens before the factory is laid out or automation is purchased.

The strongest implementations began with an 80/20-informed Strategic Intent: what growth must be created, which products and customers matter most, what must be protected, and what economic outcome the factory must deliver. Leaders then used a clean-sheet lens to design the future operation rather than simply copy the current one.

Three principles stood out:

Simplification comes before scale. Remove unnecessary variation, handoffs, queues and work before adding capacity.

Standardization matters wherever it protects quality, flow, safety, cost and the customer promise. Standardize the core; make exceptions visible and intentional.

The most critical products and customers must be protected first. A Model Factory is not an abstract production concept. It exists to protect the work that matters most to the business and its customers.

This is not a prescription to make every factory identical. It is a discipline: standardize the core where the economics and flow justify it, preserve local variation only where it creates customer value or responds to a real constraint, and make the choice explicit.

That perspective is why Front-to-Back 80/20™ cannot be an afterthought or a slogan. It is the lens that determines what the Model Factory is built to protect—and what the future organization, material system, automation and capability must make possible.

05

04 — From Material Handler to Material Flow Technician

The impact of a Model Factory on material handling is not determined by the product name. It is determined by the flow: frequency, variety, size, weight, sequence, replenishment time, line-side space, safety exposure, distance, inventory accuracy and the cost of starving the constraint.

That makes material handling a model-factory design question, regardless of whether the operation makes fasteners, enclosures, weldments, assemblies or another product entirely.

Consider one familiar as-is job: Material Handler.

In a conventional environment, the role may spend much of the day locating inventory, driving a forklift, staging material, scanning transactions, replenishing points of use and physically transporting loads.

A clean-sheet design asks which parts of that work should be removed, centralized, prepared, automated, inlined or retained. Centralize what benefits from scale and control: bulk receiving, common-component preparation, kitting, sequencing and replenishment preparation. Inline what protects takt, quality and flow: high-frequency parts, final presentation, point-of-use consumables and material where a delay immediately starves the next step. Use supermarkets, milk runs, kits or sequencing when the flow characteristics justify them.

The answer is usually a designed combination—not “central feed” by default and not “everything at the line” by default.

In a highly automated Model Factory, some movement can be performed or orchestrated by automated storage and retrieval, autonomous mobile robots, conveyors, digital replenishment signals and integrated warehouse/manufacturing systems.

The human role does not necessarily disappear. The work changes.

06

05 — Automation changes the work

Automation should follow clean-sheet process and material-flow design. It should not be used to preserve a broken sequence, automate waste or make the current organization chart permanent.

Automation can change physical movement, information movement, decisions, inspection, replenishment and escalation. Each proposal should answer five questions:

What constraint or repeated loss is being removed?

What becomes the new standard work?

What exceptions will remain, and who owns them?

What role, job, skill and decision-right changes follow?

How will the organization prove that the new system is safe, stable and Fully Productive?

The visual image of the future factory should not be an empty building filled with robots.

A more useful image is a highly automated manufacturing environment with people at the controls—not manually directing every transaction, but understanding the system well enough to see abnormal conditions, manage exceptions, protect quality, make decisions and improve flow.

Automation removes some work. It creates other work. Material handlers may become material-flow operators or technicians. Operators may become cell owners. Maintenance, Quality, Engineering, Information Technology and Operations may inherit new decisions and new failure modes.

The future job therefore cannot be created by simply adding “automation experience preferred” to today's job description.

Process, RASCI, role, job, systems, training and measures have to describe the same future operating model.

07

06 — Don't send Talent Acquisition the old factory

This has a direct workforce implication.

Talent Acquisition should not receive a request to hire 50 people until the Model Factory has determined whether the future operation actually requires 50 people doing today's jobs.

Demand should first be translated into required productive capability through the clean-sheet model-factory design.

That requires answering:

What output is required, by what Required Capacity Date and at what economics?

Which 20% of products, product families and value streams creates most of the value and must be designed first?

What work produces that output today, and what work should exist in the future?

What work should be eliminated, simplified, automated, centralized, inlined or redesigned?

What material and information flow does the future process require?

What roles, jobs and organization does the future process require—regardless of today's titles?

What capability can be built internally, and what must be recruited, redeployed, borrowed, partnered or otherwise sourced?

How will the organization know that the capability is Fully Productive?

Hiring is one mechanism inside the architecture. It is not the architecture. Talent Acquisition should receive a future-work requirement, not a larger copy of the as-is organization.

Hiring is one mechanism inside the architecture. It is not the architecture.

08

07 — Designing the Model Factory through the Compass architecture

FOUNDATION™ — Establish Strategic Intent: the next growth curve, customer outcome, demand, mix, economics, standards and Required Capacity Date. Create the clean-sheet design brief before accepting the current organization, layout or job titles as the answer.

SIMPLIFY™ — Use Front-to-Back 80/20™ to identify the Critical Few products, product families and value streams. Remove complexity and non-value-added activity before automating it. Design the future process, material flow and information flow around the model pattern.

LEAD™ — Design the future work and organization. Connect process, systems, RASCI, roles, jobs, automation ownership, material-flow ownership, Change Impact, Change & Transformation™ and Program & Project Management™. Make decision rights explicit for normal flow and exceptions.

GROW™ — Build and qualify the capability required by the future operation. Train, coach, certify, measure adoption and prove Fully Productive performance. Treat role transition as part of the design, not as an afterthought once equipment arrives.

ARC™ — Analyze reality, Refine the design and Commit to the next governed decision as learning emerges.

The architecture is not another layer on top of the Model Factory. It is how the organization keeps Strategic Intent, the economically important work, process, material system, technology, roles and capability describing the same future.

09

08 — The Model Factory is a future-work design project

A Model Factory can contain remarkable technology and still fail to produce the next growth curve.

Technology alone does not define ownership. It does not establish decision rights. It does not qualify a person. It does not create trust in a new process. It does not determine whether an exception belongs to Operations, Maintenance, Quality, Engineering or Information Technology. It does not prove adoption.

Those elements must be architected.

The model-factory validation test is therefore explicit:

Strategic Intent and the Required Capacity Date are clear.

The Critical Few product or value-stream pattern is supported by 80/20 evidence.

The clean-sheet process is defined before automation is selected.

Central, line-side, kitted, sequenced and automated material flows are chosen from the actual flow characteristics.

As-is roles are translated into future work, skills, decision rights and reporting relationships.

The future organization can run the process without relying on heroes.

A pilot proves safety, quality, flow, capacity, material availability and adoption.

The results are repeatable enough to become a model, but not copied into a product or flow with different facts.

The most important Model Factory question therefore may not be, “What technology should we buy?”

It may be:

“What work should exist when we are done—and how will we know the new system is producing the required capability?”

Once that is clear, the organization can design the process, technology, material system, roles, jobs, capability and Talent Activation system required to make the future operation real.

The most important Model Factory question may not be what technology to buy. It may be: what work should exist when we are done?