ExperienceCurve

Unit cost degression due to gaining experience

Diagram

Information

This information is part of the Business Simulation Library (BSL). Please support this work and ► donate.

The output y describes the input that is needed to produce one unit. The experience curve describes the often observed empirical trend, that with every doubling in cumulated production, the effort or cost needed per unit (e.g., unit cost) will shrink by a constant percentage (learningRate), see Louwen and Subtil Lacerda [27] for an introduction.

The output y is obtained by the following formula:

Formula.svg


Notes


  • If hasCumulatedInput = true the input u is assumed to be the cumulated value of production (N) and the initial production (N_0) is initialized with the initial value of u (unless u < 1, in which case the initial production is initialized to 1).

  • If hasCumulatedInput = false the input u is assumed to be the current rate of production which will be accumulated starting with N_0 = 1.

  • Optionally, a limit for the output value (costLimit) can be given. It will be either a floor (learningRate >= 0) or a ceiling (learningRate < 0).
  • .

Parameters (5)

initialCost

Value: 1

Type: OutputType

Description: Inial value of cost, time, defects etc. being affected by experience

learningRate

Value: 0.2

Type: Dimensionless

Description: Learning rate (LR) stating the relative reduction for each doubling of cumulated production

costLimit

Value: 0

Type: OutputType

Description: Minimum or maximum cost level (depending on sign of learning rate), which is not to be exceeded

hasCumulatedInput

Value: false

Type: Boolean

Description: = true, if input u is cumulated production value; false, if u is the current production value

hasLimit

Value: false

Type: Boolean

Description: = true, if output y will not be below the given limit or exceed it

Connectors (2)

u

Type: RealInput

Description: Input of output the cumulation of which triggeres learning

y

Type: RealOutput

Description: Costs, productivity, defects etc. being affected by experience

Components (14)

modelSettings

Type: ModelSettings

Description: Setting important global variables within a model

cumulatedProduction

Type: InformationLevel

Description: Accumulated output driving learning

producing

Type: Growth

Description: Output generation

normalizedOuput

Type: Division

Description: Division without protection against division by zero

referenceProduction

Type: ConstantConverter

Description: Reference level of output

effectLearning

Type: Power

Description: Total effect of experience

cumulatedOutput

Type: Max

Description: Cumulated output driving learning if input is already cumulated value

log2ProgressRatio

Type: ConstantConverter

Description: Binary logarithm of progress ratio

actualCosts

Type: Product_2

Description: Actual level for influenced variable

referenceCost

Type: ConstantConverter

Description: Initial level of influenced variable

minCumProd

Type: ConstantConverter

Description: Minimum cumulated production

unchanged

Type: PassThrough

Description: Output is identical to input

ceiling

Type: Clip

Description: Do not exceed maximum value

floor

Type: Clip

Description: Do not go below minimum value

Revisions

  • Introduced in v2.1.0.