Calculate production time per unit using the learning curve formula: T_n = T_1 × n^b. Estimate how labor hours decrease as cumulative output grows.
The learning curve is one of the most reliable phenomena in manufacturing: as workers repeat a task, the time per unit decreases in a predictable pattern. Specifically, each time cumulative production doubles, the time per unit drops by a constant percentage — the learning rate.
The formula is T_n = T_1 × n^b, where T_n is the time for the nth unit, T_1 is the time for the first unit, n is the cumulative unit number, and b = ln(learning rate) / ln(2). For an 80% learning curve, b = -0.322, meaning each doubling reduces time per unit to 80% of the previous level.
This calculator applies the unit learning curve (Crawford model). Enter the time for the first unit, the learning rate, and a target unit number. The calculator shows the predicted time for that unit, the cumulative time for all units, and the cumulative average time per unit.
Learning curves are essential for bidding on contracts, estimating labor budgets, forecasting production costs, and planning workforce requirements. Without learning curve projections, labor estimates will be too high for later units and too low for early units. This quantitative approach replaces subjective estimates with hard data, enabling confident planning decisions and more effective resource allocation across production operations.
T_n = T_1 × n^b Where b = ln(Learning Rate) / ln(2) Cumulative Time ≈ T_1 × Σ(i^b) for i = 1 to n Cumulative Average = Cumulative Time / n
Result: Unit 100 = 2.27 hrs, Cumulative avg = 3.48 hrs
b = ln(0.80) / ln(2) = -0.322. Time for unit 100 = 10 × 100^(-0.322) = 10 × 0.2271 = 2.27 hours. This is a 77% reduction from the first unit. The cumulative average across 100 units is approximately 3.48 hours per unit.
T.P. Wright published the first learning curve paper in 1936 based on aircraft manufacturing data. He observed that the labor hours per airplane decreased consistently as more planes were built. This observation has since been validated across virtually every manufacturing industry.
The learning curve focuses on direct labor improvement. The experience curve (Boston Consulting Group) is broader, encompassing all costs including materials, overhead, and capital. The experience curve shows that total unit cost declines 20-30% each time cumulative volume doubles.
Beyond cost estimation, learning curves help in workforce planning (how many workers needed as efficiency improves), capacity planning (output increases over time on the same equipment), and competitive analysis (experienced competitors have cost advantages that new entrants must overcome).
The unit curve (Crawford) predicts the time for a specific unit number. The cumulative average curve (Wright) predicts the average time across all units produced. Both use the same formula structure but interpret results differently.
Manual labor content (more manual = steeper learning), product complexity, workforce experience, management practices, tooling improvements, and engineering changes all affect the learning rate. Keeping detailed records of these calculations will streamline future planning and make it easier to track changes over time.
Automated processes show much flatter learning curves (90-98%) because machines don't learn the same way people do. The learning that occurs is from setup optimization, programming improvements, and process tweaks.
Eventually, further improvement is limited by machine speed, material constraints, or minimum process requirements. Most operations see diminishing learning effects after several hundred to several thousand units, depending on complexity.
Production breaks cause partial forgetting. The longer the break, the more learning is lost. A common rule of thumb: after a 3-month break, expect to restart at approximately 90-95% of where you left off on the curve.
Customers often negotiate price reductions for repeat orders based on expected learning. The learning curve justifies lower prices on later lots. Accurate learning rate data is critical for profitable contract pricing.