A Lean TPS Assembly Line Example: Manpower and Capacity Planning

Assembly line of forklifts in production, representing manpower and capacity planning in a Lean TPS environment.
A practical example of manpower and capacity planning using Lean TPS. Learn how Toyota’s approach balances efficiency, utilization, and workload to meet demand without overburdening people or processes.

Reclaiming Toyota Production System: My Lean TPS Basic Thinking

Meeting production demand requires more than adding labor. It requires structured capacity planning grounded in Lean TPS principles. Toyota’s approach integrates manpower balance, standard work, and takt alignment to ensure production meets customer demand efficiently without overburdening people or processes.

This example demonstrates how to calculate manpower requirements, assess capacity, and balance workloads using Lean TPS logic.


Step 1: Determining Manpower Requirements

A work center is tasked with producing 500 forklifts per week, each requiring 15 minutes of labor. The total weekly labor demand is:

500 units × 15 minutes = 7,500 minutes (125 hours).

The work center operates 40 hours per week at 80% utilization and 110% efficiency. The rated capacity per assembler is therefore:

40 × 0.8 × 1.1 = 35.2 available hours per assembler.

To determine manpower:
125 total hours ÷ 35.2 available hours = 3.55 assemblers.
Since partial assignments are not possible, 4 assemblers are required to meet demand.

This ensures production stability without excessive overtime or underutilization.


Step 2: Assessing Actual Production Capacity

Once manpower is assigned, the next step is to evaluate the true production capacity.

4 assemblers × 40 hours per week = 160 total hours.
Applying utilization and efficiency:
160 × 0.8 × 1.1 = 140.8 available hours.

Converting to minutes:
140.8 × 60 = 8,448 minutes.

Each unit requires 15 minutes, so actual output capacity is:
8,448 ÷ 15 = 563 units per week.

This capacity exceeds the 500-unit requirement, providing flexibility to handle variation, rework, or unplanned downtime.


Step 3: Balancing the Work Center Load

Manpower planning is only part of the equation. Lean TPS emphasizes balance across all work centers to eliminate unevenness (Mura) and overburden (Muri).

Example – Weekly Load by Product Type

  • Model A: Setup 3 hours, run time 2 hours per piece.
  • Model B: Setup 5 hours, run time 1 hour per piece.

A three-week workload review reveals:

  • Week 1 = 43 hours (3 hours over capacity)
  • Week 2 = 38 hours (2 hours under capacity)
  • Week 3 = 41 hours (1 hour over capacity)

Total = 122 hours, or 2 hours over capacity.

By identifying these imbalances, production planning can redistribute work, adjust setups, or improve sequence planning to maintain flow.


Step 4: Applying Lean TPS Thinking

Toyota’s manpower and capacity planning process is more than arithmetic. It is a visual, data-driven system that supports stability, flow, and people development. The focus is to:

  • Balance workloads across shifts and lines.
  • Maintain flexibility without creating excess labor.
  • Improve methods before increasing manpower.
  • Use Standardized Work to stabilize repetitive tasks.

The result is a system that meets customer demand while preventing overproduction, underutilization, and burnout.

Lean TPS ensures that capacity planning supports both performance and respect for people. It aligns manpower, process capability, and continuous improvement into one consistent management system.

Lean TPS House diagram showing Just In Time, Jidoka, Heijunka, Standardized Work, and Kaizen positioned within the Toyota Production System architecture
This Lean TPS Basic Training visual explains how Kaizen operates within the governed architecture of the Toyota Production System. Just In Time and Jidoka function as structural pillars, Heijunka and Standardized Work provide stability, and Kaizen strengthens the system only when standards and control are in place. The image reinforces
Lean TPS Swiss Cheese Model showing how governance failures propagate from organizational systems to gemba outcomes, and how TPS prevents conflicts that Theory of Constraints resolves downstream.
Theory of Constraints manages conflict after instability forms. Lean TPS prevents conflict through governance of demand, capacity, and Quality before execution begins.
Takahama Line 2 Andon board showing real time production status and Quality control in the Toyota Production System
Dashboards and scorecards increase visibility, but they do not govern work. In Lean TPS, Andon exists to control abnormality in real time by enforcing stop authority, response timing, and leadership obligation to protect Quality.
Lean TPS Disruptive SWOT transforms traditional SWOT from a static listing exercise into a governed leadership system. Through Survey, Prioritize, and Action, it aligns strategic direction with Quality, system stability, and explicit leadership obligation within a Lean TPS governance framework.
Balance scale showing Respect for People and Continuous Improvement grounded in Quality governance within Lean TPS.
In Lean TPS, Respect for People and Continuous Improvement are not independent goals. Both emerge from Quality governance, where leaders define normal work, make abnormality visible, and respond to protect system stability.
Lean TPS shop floor before and after 5S Thinking showing visual stability that enables problem detection and problem solving
5S Thinking is not about making the workplace look clean or impressive. In Lean TPS, it functions as a visual reset that restores the ability to see normal versus abnormal conditions. When the environment is stabilized, problems surface quickly, Quality risks are exposed earlier, and problem solving becomes possible at