Module 7 - Learning Curve Analysis Flashcards

(90 cards)

1
Q

What is the main concept behind the learning curve?

A

Cost improvement, reduction of cost with the production of multiple units

Cost improvement occurs when units are produced under the same processes by a consistent workforce without undue delays.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What method is commonly used to quantify the degree of cost improvement?

A

Learning curve analysis

This method can also predict a learning curve for future production units based on historical data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

According to learning curve theory, cost improvement manifests as a constant percent reduction for each _______.

A

doubling in quantity

This principle applies to various costs, including labor and material costs.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are the two types of costs discussed in the context of learning curves?

A
  • Individual unit cost (Unit Learning Curve Theory)
  • Cumulative Average Unit Cost (Cumulative Average Theory)

These costs can be calculated from one another.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the Lot Midpoint (LMP) in learning curve analysis?

A

The equivalent unit number of the most representative point of the lot

It is necessary for applying Unit Learning Curve with lot data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

True or false: Learning curve analysis can only be applied to touch labor hours.

A

FALSE

Learning curves have been applied to labor cost, material cost, and other elements.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is the Learning Curve Slope (LCS)?

A

The percentage by which cost decreases as the production units double

It is a key metric in learning curve analysis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is the formula for the learning curve?

A

Y = aX^b

This equation can represent either individual unit cost or cumulative average unit cost.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What does the variable b represent in the learning curve equation?

A

The actual slope of the line formed when the learning curve data is transformed into log space

It relates to the Learning Curve Slope (LCS).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What happens to the learning curve when data is examined in log space?

A

The points form a straight line

Logarithmic transformations help with plotting data and performing calculations.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the relationship between b and LCS?

A

b = ln(LCS) / ln(2)

This shows how the learning curve slope is derived from the logarithmic transformation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Define CUMAV data.

A

Cumulative unit number and cumulative average unit cost of units 1 through k

CUMAV data is used in cumulative average learning curve theory.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the significance of normalizing data in learning curve analysis?

A

To remove exogenous effects such as labor force skill level and product changes

Normalization helps in accurately analyzing learning curves.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What does the term Cost Improvement Curve (CIC) refer to?

A

Another name for the learning curve concept

It emphasizes the broader applications of cost improvement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the first unit cost of unit 2 if the unit cost of unit 1 is $1,000 and the LCS is 90%?

A

$850

Calculated as 0.85 * $1,000.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What are the two competing learning curve theories?

A
  • Unit Learning Curve (ULC)
  • Cumulative Average (CUMAV)

These theories describe how individual unit costs and cumulative average unit costs follow the learning equation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is the purpose of normalizing data in learning curve analysis?

A
  • Remove effects due to labor force composition
  • Track changes in end-item content
  • Address irregularities in production timing

Normalization improves prediction accuracy by eliminating noise in the data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

True or false: Fitting a learning curve to unnormalized data will always result in high prediction accuracy.

A

FALSE

While the model may be statistically significant, prediction accuracy is often lowered without normalization.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are the four conditions under which analysts should utilize learning curve methodology?

A
  • Direct labor comprises the majority of total price
  • Contract price is high enough for data collection
  • Duration of production is long enough for sufficient data
  • Item in production is nonstandard

These conditions ensure the effectiveness of learning curve analysis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Who first asserted the CUMAV learning curve theory?

A

T.P. Wright in 1936

The theory is also known as Wright Curves and predicts learning curve effects in an aggregate manner.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What does the standard learning curve equation represent?

A

Y = aX^b

Where Y is the cumulative average cost, X is the quantity of units produced, a is the theoretical first unit cost, and b captures cost improvement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is the theoretical first unit cost (T1) in the learning curve equation?

A

A critical component representing the cost of the first unit

It is often not the actual cost of the first unit in practice.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is the learning curve slope (LCS) formula?

A

LCS = 2^b

This formula captures the percentage of cost improvement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is the first step in determining a learning curve?

A

Gather available data and normalize it

Normalization may include converting to constant year dollars.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
What should be plotted in **unit space** during learning curve analysis?
Unit numbers and costs ## Footnote This helps to observe potential LDA problems before transforming data.
26
What is the purpose of plotting data in **log space**?
To determine if learning curve analysis is suitable ## Footnote The data should approximate a straight line in log space.
27
What statistical method is typically used to determine the equation of the line in log space?
Ordinary Least Squares (OLS) regression ## Footnote This method provides the intercept and slope for the learning curve.
28
How do you transform log space parameters back into unit space?
Exponentiate the y-intercept value ## Footnote This gives the theoretical first unit cost (T1).
29
What is the formula to calculate the **average unit cost** of the first 20 units?
a(20)^b ## Footnote This demonstrates the dramatic effect of the learning curve on costs.
30
What is the total cost of the first 20 units calculated by multiplying the average unit cost by the number of units?
49.54 x 20 = 990.72 ## Footnote This shows how the learning curve impacts total production costs.
31
What is the formula to calculate the **cost of a single unit** or group of units?
a(20)^(b+1) - a(19)^(b+1) ## Footnote This calculates the cost of the 20th unit based on cumulative costs.
32
What does **CUMAV** stand for?
Cumulative Average Cost ## Footnote CUMAV is used to analyze the average costs associated with production units.
33
In CUMAV, how is the **cumulative average cost** of the first 20 units determined?
By adding the predicted cost of units 11-20 to the actual costs of units 1-10 ## Footnote This method allows for a comprehensive understanding of total production costs.
34
What is the **learning curve slope** used in the example?
0.85 ## Footnote This slope indicates the percentage reduction in cost with each doubling of production.
35
The cumulative average cost through 10 units is multiplied by the learning curve slope to find the cost of the first 20 units. What is the resulting cost?
49.54 ## Footnote This calculation reflects the expected cost reduction due to learning effects.
36
What type of data is referred to as **lot data**?
Data for entire production lots (multiple units) ## Footnote Lot data is contrasted with unit data, which refers to individual production units.
37
In the context of ULC theory, what does **Y** represent?
The cost of the Xth unit ## Footnote ULC theory focuses on individual unit costs rather than cumulative averages.
38
What does **X** represent in the ULC equation?
The sequential unit number of the unit being calculated ## Footnote This variable is crucial for determining individual unit costs.
39
What is the significance of the **log base two of the LCS** in ULC theory?
It captures cost improvement as a negative number ## Footnote This value is essential for understanding the learning curve's impact on costs.
40
What is the first step in the **ULC example** with unit data?
Collect the data ## Footnote Gathering accurate data is crucial for effective analysis in ULC theory.
41
What is the purpose of plotting data in **unit space**?
To observe potential LDA problems ## Footnote This step helps in identifying issues with the learning curve analysis.
42
What does a **poor R² value** indicate in the context of learning curve analysis?
An LDA problem may exist ## Footnote A low R² value suggests that the data does not fit well to the learning curve model.
43
What is the expected cost of the **20th production unit** in the ULC example?
To be determined using ULC theory ## Footnote The calculation involves applying the learning curve equation to individual unit costs.
44
What is the **total cost of 20 production units** calculated from ULC theory?
To be determined by summing individual unit costs ## Footnote This approach contrasts with cumulative average cost calculations.
45
What does **CAUC** stand for?
Cumulative Average Unit Cost ## Footnote CAUC is a key metric in analyzing production costs over time.
46
What is the significance of the **learning curve** in production?
It predicts cost reductions as production increases ## Footnote Understanding the learning curve helps in forecasting future production costs.
47
What does the **lot midpoint** represent in ULC theory?
The unit at which the average cost of the lot occurs ## Footnote It is not the same as the average unit number of the lot and is a single point representing the entire lot.
48
The **Average Unit Cost (AUC)** of a lot is calculated by dividing the sum of individual units by what?
The number of units (N) ## Footnote AUC = (Sum of individual units) / N
49
Fill in the blank: The formula for **lot midpoint (LMP)** is given by __________.
LMP = (Sum of i^b / N)^(1/b) ## Footnote This formula involves the exponent b twice, in the summation and outside.
50
True or false: The **lot midpoint** is simply the middle unit of the lot.
FALSE ## Footnote The lot midpoint is a single point that represents the entire lot, not necessarily the middle unit.
51
What is the relationship between **LMP** and **LCS** in ULC theory?
LMP is needed to calculate LCS, and vice versa ## Footnote This creates a dilemma in determining accurate values.
52
What are two potential solutions to the dilemma of calculating **LMP** and **LCS**?
* Determine approximations or heuristics * Use a closed-form formula for LMP involving LCS ## Footnote Both approaches can help estimate LMP without needing LCS directly.
53
The **AUC** of a lot is set equal to what in ULC theory?
The basic learning curve formula ## Footnote This allows for solving for LMP in relation to the learning curve.
54
What does the **learning curve** represent in the context of ULC?
The relationship between unit number and unit cost ## Footnote It shows how costs decrease as production increases.
55
What is the formula for calculating **LCS** based on the learning curve?
LCS = 2^b ## Footnote This is derived from the learning curve parameters.
56
What happens to the accuracy of the **LMP estimation formula** as the number of lots increases?
It becomes increasingly accurate ## Footnote Conversely, the lower the lot number, the less accurate the formula tends to be.
57
Fill in the blank: The **LMP** is the point where the horizontal line representing the average unit cost for a lot intersects the __________.
Learning curve ## Footnote This intersection represents the most representative point for that lot.
58
The **arithmetic mean** always overestimates LMP and underestimates the cost of the lot. True or False?
TRUE ## Footnote The arithmetic mean is contrasted with the geometric mean in terms of their estimations of LMP and lot costs.
59
The **geometric mean** always underestimates LMP and overestimates the cost of the lot. True or False?
TRUE ## Footnote This relationship highlights the differences in how these means calculate values.
60
True LMP always lies between the **arithmetic mean** and the **geometric mean**. True or False?
TRUE ## Footnote This statement emphasizes the positioning of true LMP in relation to the two means.
61
As the number of lots increases, the formula for estimating LMP becomes increasingly more _______.
accurate ## Footnote This indicates that larger datasets improve the reliability of the estimation.
62
The lower the lot number, especially the very first lot, the less _______ the formula becomes.
accurate ## Footnote Early lots tend to introduce more variability in estimations.
63
The variable **b** is a required input for the more accurate LMP formula. True or False?
TRUE ## Footnote The value of b is essential for refining LMP calculations.
64
The exact LMP formula involves an open-form summation with _______ as the exponent of every term.
b ## Footnote This complexity can be managed with spreadsheet tools like Excel.
65
The formula resulting from the approximation is shown in Figure _______.
7.28 ## Footnote This figure illustrates the more accurate LMP approximation formula.
66
In the ULC example, the task is to determine the cost of the 4th lot, the total cost of 20 units, and the _______ cost of 20 units.
average ## Footnote This highlights the key calculations involved in the ULC analysis.
67
The second step in the ULC example involves plotting the data using a _______ plot in unit space.
scatter ## Footnote This visual representation aids in understanding the data distribution.
68
The approximate **b**-value in the ULC example is __________.
-0.2402 ## Footnote This value is used to generate the second column calculated using the LMP formula.
69
The CUMAV theory's smoothing effect gives it a significant advantage over _______.
ULC ## Footnote This comparison highlights the strengths of different learning curve theories.
70
CUMAV is easier to use when working with large amounts of data due to its _______.
normalization and calculation steps ## Footnote This efficiency contributes to its popularity in data analysis.
71
Objections to CUMAV can be summed up as: it uses a synthetic variable, confuses the issue of true estimating error, obscures the variation of the observed variable, and produces _______.
artificially favorable statistics ## Footnote These objections highlight the limitations of the CUMAV approach.
72
The **CUMAV iterative method** can account for missing and concurrent lots. True or False?
TRUE ## Footnote This method improves the analysis by treating each data point independently.
73
The traditional CUMAV theory uses the last unit number as the representative point for each lot. True or False?
TRUE ## Footnote This methodology has its own set of issues, particularly with missing data.
74
What is the **CUMAV iterative method** used for?
To account for missing and concurrent lots ## Footnote Benefits include no data smoothing, more reliable goodness-of-fit measures, and easier identification of outliers.
75
List the **benefits** of the CUMAV iterative method.
* No data smoothing * More reliable goodness-of-fit measures * Easier identification of outliers ## Footnote This method treats every data point independently.
76
True or false: The **CUMAV iterative method** is less complicated than ULC.
FALSE ## Footnote It requires procedures at least as complicated as ULC but is less intuitive.
77
What does **ULC** stand for?
Unit Learning Curve ## Footnote ULC is a theory used in learning curve analysis.
78
What is the **initial guess** value used for ULC theory in the example?
-0.1938 ## Footnote This value was calculated using the previous LMP heuristic approach.
79
What is the purpose of **learning curve analysis**?
To determine the cost of future units or lots in production ## Footnote It extrapolates from actual production data.
80
What are the **three possible scenarios** for a new program in learning curve analysis?
* Brand-new program * Second production line * Existing production line restart ## Footnote Each scenario has different approaches to estimating costs.
81
What should be justified with thorough analysis in learning curve estimating?
The chosen **LCS value** ## Footnote It often has the greatest effect on the production cost estimate.
82
What is the effect of a modest change in **learning curve slope**?
It can significantly affect production costs as early as the 5th unit ## Footnote For example, a difference between 95% and 85% LCS can lead to a 23% cost difference.
83
What is the **Quantity As an Independent Variable (QAIV)** approach?
Includes production quantity as an additional independent variable in CER development ## Footnote This method estimates the production cost for the entire production run.
84
What is the **average unit cost (AUC)** calculated from?
The lot cost divided by the number of units in the lot ## Footnote AUC is not the same as CAUC.
85
What is the **CUMAV** theory known for?
It provides a tighter fit for regression statistics compared to ULC ## Footnote CUMAV is chosen based on higher R² values.
86
What does the **goodness-of-fit** measure indicate?
How well a statistical model fits the data ## Footnote It is crucial for validating the chosen learning curve theory.
87
What is the **impact of automation** on learning curves?
Less learning is expected with high automation and little touch labor ## Footnote The fastest learning corresponds to a steep curve.
88
What are some **factors affecting slope** in learning curves?
* Complexity of manufacturing process * Initial skill level of workers * Degree of technology * Types of resources consumed * Involvement of unions ## Footnote These factors can influence the steepness or flatness of the learning curve.
89
What is the **significance of regression statistics** in learning curve analysis?
They confirm the statistical significance of the slope and intercept ## Footnote Both theories must be reviewed for their fit to the data.
90
What is the **cost estimation approach** for follow units in learning curves?
Use a steady-state should-cost method run back up the learning curve ## Footnote This helps estimate earlier unit costs based on later units.