Can You Do Lean Manufacturing in a Job Shop? Yes — Here's How to Adapt It

Workers reviewing production schedules on a job shop floor with diverse parts in progress

Walk into any job shop and mention Toyota Production System and you'll hear a variation of the same objection: "That's great for making millions of identical cars. We make 3,000 different part numbers and almost nothing repeats. Lean doesn't apply here."

It's a reasonable objection — and it's wrong. Not completely wrong. The specific tools Toyota developed for high-volume repetitive production don't transfer unchanged to a high-mix, low-volume (HMLV) environment. But the underlying principles — eliminate waste, expose problems, continuously improve — apply everywhere manufacturing happens.

After 35 years of helping job shops, machine shops, fabricators, and contract manufacturers schedule and improve their operations, User Solutions has seen lean work in environments with 50,000 active part numbers and 2-week average job durations. The key is knowing what to take directly from Toyota, what to adapt, and what to leave behind entirely.

Why Job Shop Owners Dismiss Lean (and Why That's Costly)

The classic lean toolkit was developed at Toyota in Nagoya in the 1950s through 1970s. Toyota's production system was designed around a specific context: making the same car models in predictable quantities for years at a time. Under those conditions, you can calculate takt time (production pace tied to customer demand), implement one-piece flow (move one unit at a time between identical process steps), and use replenishment kanban (restock signals tied to consumption of standard parts).

A job shop making custom aerospace brackets, hydraulic manifolds, and medical enclosures has almost nothing in common with that context. Every order is different. Demand is lumpy and hard to predict. Setups are frequent and expensive. The "flow" between operations is different for every job.

The mistake job shop owners make is rejecting lean because its Toyota-specific tools don't fit — and therefore missing the tools that do fit. The cost of that rejection is operating a shop with far more waste than necessary: tool search time, machine downtime from poor scheduling, quality rework from rushed setups, late deliveries from queue mismanagement.

Let's separate what transfers directly from what needs adaptation.

What Transfers Directly from Lean to Job Shops

5S: Workplace Organization

5S (Sort, Set in Order, Shine, Standardize, Sustain) is the most universal lean tool. It requires no modification for a job shop. The benefits — reduced tool search time, fewer errors from misplaced items, better safety, cleaner work environment — apply regardless of how repetitive the work is.

In a job shop with 50+ setups per day across a dozen work centers, 5S typically saves 15–30 minutes per operator per shift simply by eliminating time spent searching for tooling, fixtures, and documentation. On a 20-person shop, that's 5–10 hours of recovered capacity per day — the equivalent of hiring another operator.

The key to sustaining 5S in a job shop is the "Set in Order" step. Every tool, fixture, gauge, and consumable needs a designated location with a visual indicator (shadow board, tape outline, label). When a job requires a non-standard fixture, it has a temporary home for the duration of that job and returns to storage when complete.

Standard Work for Setups

Standard work means documenting the one best-known way to do a task so that task can be done consistently every time. In repetitive production, standard work covers the production cycle. In a job shop, the highest-leverage application is setup standardization.

Most job shops have tribal knowledge about how to set up each machine type, but that knowledge lives in individual operators' heads. When the experienced operator calls in sick, setup time doubles and first-part quality drops. Documented setup procedures — with photos, tooling specifications, first-piece inspection steps, and expected setup times — convert tribal knowledge into organizational capability.

Standard setup procedures also make training dramatically faster. New operators can reach 80% proficiency in weeks rather than months when they have clear documentation to follow and experienced operators can focus on coaching edge cases rather than teaching basic setup from scratch.

SMED: Quick Changeover

SMED (Single-Minute Exchange of Die) is arguably the highest-ROI lean tool for most job shops. SMED is the systematic reduction of setup time — the time between the last good part of one job and the first good part of the next.

The SMED methodology works in four steps:

Record current setup through video or detailed observation
Separate internal activities (machine must be stopped) from external activities (can be done while machine runs)
Convert as many internal activities as possible to external
Streamline all remaining activities

In a typical job shop, 40–60% of setup time involves activities that can be moved to external: gathering tooling and documentation, pre-staging fixtures, programming and proving out CNC code, setting up inspection equipment. Moving these activities outside the machine stop reduces setup time without purchasing a single new tool.

A job shop running 200 setups per month at an average of 90 minutes each is consuming 300 machine-hours in setup. A 40% SMED reduction — completely achievable in most shops — recovers 120 hours of capacity per month. At $80/hour machine rate, that's $9,600 in recovered capacity every month, from process improvement alone.

Visual Management

Visual management means making the state of the operation visible at a glance: what's running, what's waiting, what's blocked, what's late. In Toyota plants, this includes production boards, andon lights, kanban cards, and takt time clocks. In a job shop, the implementations differ but the need is identical.

Effective job shop visual management includes:

Job travelers that move with each job and show status at each operation
Work-in-process boards at each work center showing jobs queued, in process, and complete
Machine status indicators (running, setup, down, idle)
Hot job markers identifying expedited or at-risk orders
Schedule vs. actual boards tracking on-time completion by shift or day

The principle is the same as Toyota: if you have to ask what's happening, the system is failing. The status should be visible without a conversation.

Gemba Walks and Continuous Improvement Culture

Gemba ("the real place" in Japanese) walks — management going to the shop floor regularly to observe, ask questions, and identify improvement opportunities — require no modification for a job shop. The discipline of going to where work happens, seeing waste directly, and engaging operators in problem-solving is universally applicable.

In a job shop, gemba walks are particularly valuable for catching scheduling problems early. An experienced scheduler or plant manager walking the floor can see a work center accumulating queue while a downstream machine sits idle — a capacity imbalance that doesn't show up in reports until it becomes a late delivery.

What Requires Adaptation

Takt Time → Finite Capacity Scheduling

Takt time is the heartbeat of a lean line: available time divided by customer demand. If a plant runs 480 minutes per day and customers want 240 units, takt time is 2 minutes. Every process step is designed to complete in 2 minutes or less.

In a job shop, this concept breaks down because demand is not a predictable rate of identical units. You might have 47 different jobs due this week, each with different routing, different operation times, and different due dates.

The lean equivalent for job shops is finite capacity scheduling: planning each unique job against available machine capacity, respecting actual constraints, and generating a schedule that shows what should be running on what machine at what time. Done well, finite capacity scheduling eliminates the same waste that takt time eliminates in repetitive production — it just uses a fundamentally different algorithm.

Tools like RMDB apply this logic to job shops, modeling each machine's capacity, each job's routing and operation times, and generating schedules that maximize throughput while meeting due dates. This is lean scheduling adapted for HMLV reality.

One-Piece Flow → Minimize Batch Size

One-piece flow — moving one unit at a time between operations — is the lean ideal that eliminates queue time between process steps. In a job shop, true one-piece flow is rarely achievable because operation times vary widely between steps and setups are costly.

The adaptation is minimum viable batch size: reduce batch sizes as far as setup economics allow. If a CNC setup takes 45 minutes and cycle time per part is 3 minutes, running batches of 1 is uneconomical. But running batches of 500 when a customer wants 50 creates unnecessary WIP and lead time. The right batch size balances setup amortization against queue time and WIP cost.

As SMED reduces setup time, the economic batch size shrinks. A setup that once required 90 minutes to justify batches of 100 now requires only 30 minutes — and batches of 30 become economical. Lead time compresses accordingly.

Replenishment Kanban → Order-Based Pull

Kanban in Toyota's system is a replenishment signal: when a downstream process consumes a container of parts, the card triggers upstream production of a replacement container. This works when you're making the same parts repeatedly.

In a job shop making custom parts, there's no standard part to replenish. The adaptation is order-based pull: work is released to the shop floor only when capacity is genuinely available to process it, not as soon as the order is received. This is the "pull" principle without the kanban card mechanism.

In practice, this means resisting the temptation to release every new order immediately. Releasing too many jobs simultaneously creates large queues at bottleneck machines, hides the real due-date risk, and makes it impossible for operators to know what to work on. Releasing jobs sequenced to available capacity keeps WIP manageable and makes priorities clear.

Cellular Manufacturing: The Bridge Between Lean and Job Shop

The most powerful structural adaptation for applying lean in a job shop is cellular manufacturing — grouping machines and operators into dedicated cells focused on a product family.

Most job shops, when they analyze their order data, find that 20% of their part families represent 60–80% of their volume. A shop making 3,000 part numbers might find that 400 part numbers in 5 product families account for two-thirds of all revenue. Those 400 parts have enough commonality in routing and operation types to justify dedicated cells.

Within a dedicated cell, the conditions that make Toyota's lean tools work are recreated: consistent routing, predictable cycle times, defined takt time, and the possibility of genuine flow. The cell can use standard lean tools unchanged. The remaining one-off jobs route through a general-purpose area where finite capacity scheduling governs priorities.

This hybrid approach — lean cells for the repetitive portion, capacity scheduling for the custom portion — captures the best of both worlds. Many job shops achieve 50–70% of their volume through cells while maintaining full flexibility for the custom work that defines their value proposition.

A Practical Implementation Path for a 10–50 Person Job Shop

For a job shop owner or plant manager starting lean for the first time, here is a realistic 12-month path:

Months 1–3: Foundation

Run 5S events at highest-traffic work centers
Document top 10 setups with standard work sheets and photos
Begin weekly gemba walks with a structured observation checklist

Months 4–6: SMED

Select the 3 highest-frequency setups and run SMED events
Target 40% reduction in setup time for those setups
Measure and post setup time data visibly at each machine

Months 7–9: Scheduling

Implement finite capacity scheduling for the full shop
Establish work-in-process limits at each work center
Create visual job status boards at every key work center

Months 10–12: Cellular Evaluation

Analyze order data to identify candidate product families for cells
Run a pilot cell for the top candidate family
Measure lead time and quality before and after

This sequence works because each phase builds on the previous one. You can't sustain cells without good 5S and standard work. Finite capacity scheduling is more effective when setups are predictable through SMED. And cells are worth evaluating only once you understand your flow patterns through scheduling data.

The Bottom Line: Lean Applies, Differently

The job shop owner who says "lean doesn't apply here" is half right — Toyota's specific implementations don't apply unchanged. But the person running that shop without 5S, without SMED, without visual management, without finite capacity scheduling, is running a shop full of waste that continuous improvement could eliminate.

Lean manufacturing is not a Toyota-specific methodology. It's an approach to systematically identifying and eliminating waste. Every job shop has waste — in setup time, in tool search, in expediting, in late deliveries, in quality rework. The tools to eliminate that waste exist and have been proven in hundreds of HMLV shops over decades.

The adaptation required is real. The payoff is larger in a job shop than in a repetitive plant, precisely because the starting level of waste is higher and the margin for improvement is greater.

Yes — but with deliberate adaptation. Lean principles like 5S, SMED, visual management, and continuous improvement culture transfer directly. Principles designed for high-volume repetitive production (takt time, one-piece flow, replenishment kanban) need modification for high-mix, low-volume work. The mistake is copying Toyota's tools without understanding the underlying logic.

Takt time. In repetitive production, takt time (available time ÷ customer demand) sets the rhythm of the whole plant. In a job shop where every order is different and volumes are unpredictable, you cannot rely on a single takt number. Instead, job shops use finite capacity scheduling — planning each unique job against available machine hours — which achieves the same waste-elimination goal through a different mechanism.

Cellular manufacturing groups machines and operators dedicated to a product family — for example, all small turned parts, or all sheet metal enclosures. Within the cell, flow is predictable and lean tools apply normally. The cell is a lean island inside a job shop ocean. Many job shops find that 60–80% of their volume concentrates in 20% of their part families, which is enough to justify 2–4 cells while keeping a general-purpose area for true one-offs.

SMED applies directly and often delivers the highest ROI of any lean tool in a job shop. Because job shops constantly switch between orders, setup time is one of the biggest sources of capacity loss. Reducing setup time by 50% doesn't just save labor — it makes smaller batch sizes economically viable, which compresses lead time across every order in the queue. In our experience with job shops running 200+ setups per month, SMED consistently frees 15–25% of spindle capacity.

Ready to bring lean scheduling to your job shop? Contact User Solutions to see how RMDB applies finite capacity scheduling — the lean tool built for high-mix, low-volume production. Trusted by GE, Cummins, BAE Systems, and hundreds of job shops for 35+ years.

For more on the foundations of lean manufacturing, see our lean manufacturing glossary, kanban in manufacturing, and SMED quick changeover guides.

Can You Do Lean Manufacturing in a Job Shop? Yes — Here's How to Adapt It

Why Job Shop Owners Dismiss Lean (and Why That's Costly)