# Improving the Application of the "Learning Curve"

Project Report 2017 30 Pages

## Excerpt

## Table of Contents

Table of Contents

List of figures

List of Tables

1. Introduction

1.1 Background and motivation

1.2 Problem statement

2. Literature Review

2.1 Learning Curve

2.2 Learning Curves Approaches

A. Arithmetic Approach

B. Logarithm Approach

C. Coefficient Approach

2.3 The Standard time and Normal time

3. Research Question

4. Objectives

5. Experimentation

5.1 Methodology

5.2 Apparatus

6. Observations

7. Analysis

8. Results

9. Discussions

10. Conclusion

11. References

Appendices

Appendix A: Ethical considerations

Appendix B: Limitations of Learning curve theory

Appendix C: Typical Learning Rates

Appendix D: England Flyer Instructions

## List of figures

Figure 1 Learning curve (Cumulative Average & Time)

Figure 2 Folding Instructions (England Flyer)

## List of Tables

Table 1 Example of Learning-curve effects

Table 2 Learning curve-Arithmetic Approach

Table 3 Learning Curve-Coefficient Approach

Table 4 Raw data

Table 5 Raw data (Unit & time)

Table 6 Cumulative averages

Table 7 Unit improvement factor

Table 8 Learning rate

## Abstract

There exist some fields in which the learning curves can be applied; they have a much wider applications including manufacturing and marketing strategy. However, they underlay the concept of continuous improvement, pricing decisions, work scheduling, standard setting, direct labour budget etc. Learning curve states that decreasing man hours are required to accomplish any repetitive task as the operation is continued; then knowledge of learning curve can be useful both in planning and control.

Estimates of learning curves are used in many applications in organizations and they can be effective with the job which is repetitive in nature particulars with same machinery and tools; thus this study aims at improving the application of the learning curve which describes how knowledge is acquired and retained when people repeat a process. The overall purpose of this report is to determine whether or not the learning curve theory can be replicated.

Data will be gathered based on the England flyer folding instructions; thereafter we will be plotting the learning curve by calculating the cumulative average, estimating the table of unit improvement curves and determining the learning rate.

## 1. Introduction

### 1.1 Background and motivation

When a new product or process is started, performance of worker is not at its best and learning phenomenon takes place. As the experience is gained, the performance of worker improves, time taken per unit reduces; and then his productivity goes up. Thus, people who learn always increase their experience, they gain skill and ability from their own experience and the results throughout processes are improved. The usual use of learning curves is to estimate the labor hours and thereby much of the cost, of a manufactured good that is built in significant quantities.It can be applied to the individuals or organisations based on several assumptions:

- Standardised product: the product isn’t changing; then every time the product changes, the learning effect will stop and start again.

- Little or no breaks in production & little or no labour turnover.

- The unit time will decrease at a decreasing rate.

- Complex operation: simple operation doesn’t require learning.

- The reduction in time will follow a predictable pattern.

- The process is labour intensive: the labour determines the speed of the process.

The learning curve is a line displaying the relationship between *unit production time* and *the cumulative number of units produced*. Learning curves have been applied to creation of documents, boring of tunnels, drilling of wells, upgrades of previously manufactured products, and many other repetitive activities This time around, two learning curve models are in widespread use: the “unit” (U) model, due to Crawford, and the “cumulative average” (CA) model, due to Wright. Some manufacturing companies also apply “learning” to the purchase of raw materials and also to the purchase of manufactured components from other companies (Evin Stump P.E). Then, an organization also acquires knowledge in its technology, its structure, documents that it retains, and standard operating procedures [1].

Applications of the learning curve:

**- Internal**: determine labor standards and rates of material supply required.

**- External**: determine purchase costs.

**- Strategic**: determine volume-cost changes.

### 1.2 Problem statement

For purposes of the current study, an assignment has been set to determine whether or not the learning curve theory can be replicated.

## 2. Literature Review

The method study was originally designed for the *analysis* and *improvement of repetitive manual work*; though it can be used for all types of activity at all levels of an organisation, is the process of subjecting work to systematic, critical scrutiny to make it more effective and or more efficient. The aim of method study is to *analyse a situation*, *examine the objectives of the situation* and then *to synthesize an improved*, *more efficient and effective method or system* [2].

In fact, it is one of the keys to achieving productivity improvement whereby the process is often seen as linear. The basic procedure was first developed and articulated by Russell Currie at Imperial Chemical Industries (ICI) and consists of six steps (SREDIM):

- Select (the work to be studied);

- Record (all relevant information about that work);

- Examine (the recorded information);

- Develop (an improved way of doing things);

- Install (the new method as standard practice);

- Maintain (the new standard proactive).

### 2.1 Learning Curve

The learning curve is based on a *doubling of production*: That is, when production doubles, the decrease in time per unit affects the rate of the learning curve. Mostly, cost is related to time or labor hours consumed, learning curves are very important in industrial cost analysis. A key idea underlying the theory is that every time the production quantity doubles, we can expect a more or less fixed percentage decrease in the effort required to build a single unit (the Crawford theory).

The “learning effect” was first noted in the 1920s in connection with aircraft production. It use was amplified by experience in connection with aircraft production in WW II. Initially, it was thought to be solely due to the learning of the workers as they repeated their tasks. Later, it was observed that other factors probably entered in, such as improved tools and working conditions, and various management initiatives. Learning curve was first described by psychologist Hermann Ebbinghaus in 1885 and is used as a way to measure production efficiency and to forecast costs. Afterward, the learning curves were applied to industry in a report by T. P. Wright of Curtis-Wright Corp. In 1936 Wright described how direct labor costs of making a particular airplane decreased with learning, a theory since confirmed by other aircraft manufacturers. Regardless of the time needed to produce the first plane, learning curves are found to apply to various categories of air frames [3].

Furthermore, it has since been applied not only to labor but also to a wide variety of other costs, including material and purchased components; and the power of the learning curve is so significant that it plays a big role in many strategic decisions related to employment levels, costs, capacity, and pricing.

The Learning curves are mathematical models used to estimate efficiencies gained when an activity is repeated; the principle is shown as: ** T ** ×

**= Time required for the**

*Ln***th unit**

*n*Where ** T ** = unit cost or unit time of the first unit

* L* = learning curve rate

* n* = number of times

**is doubled**

*T*For instance, if the first unit of a particular product took *10 labor-hours*, and if a *70%* learning curve is present, the hours the fourth unit will take require doubling twice; from 1 to 2 to 4.

Thus, the formula is: **Hours required for unit 4 = 10 × (.7)2 = 4.9 hours**

Industry learning curves vary widely and have various effects; and different organizations, products, have different learning curves. The rate of learning varies depending on the quality of management and the potential of the process and product. Any change in *process*, *product*, *or personnel* disrupts the *learning curve*. Thus, caution should be exercised in assuming that a learning curve is continuing and permanent.

Table 1 Example of Learning-curve effects

Abbildung in dieser Leseprobe nicht enthalten

### 2.2 Learning Curves Approaches

### A. Arithmetic Approach

The arithmetic approach looks to be the simplest approach to learning-curve equations; each time that production doubles labor per unit declines by a constant factor, known as the learning rate. Thus, if we the learning rate is 80% and that the first unit produced took 100 hours, the hours required to produce the second, fourth, eighth, and sixteenth units are as follows:

Table 2 Learning curve-Arithmetic Approach

Abbildung in dieser Leseprobe nicht enthalten

The hours required to produce *N* units and *N* is one of the doubled values. But, the Arithmetic approach does not show how many hours will be needed to produce other units. For this flexibility, the logarithmic approach seems useful.

### B. Logarithm Approach

The logarithmic approach determines labor for any unit, *TN*, by the formula:

*TN* = *T* 1 (*Nb*)

Where ** TN ** = time for the

*N*th unit

* T*

**1**= hours to produce the first unit

** b ** = (log of the learning rate)/ (log 2) = slope of the learning curve

For instance, the learning rate for a particular operation is 80%, and the first unit of production took 100 hours. The hours required to produce the third unit may be computed as follows:

*TN* = *T* 1 (*Nb*)

T3 = (100 hours) (3ᵇ)

= (100) ( )

= (100) ( = 70.2 labor-hours

### C. Coefficient Approach

The learning-curve coefficient is shown by the following equation:

*TN* = *T* 1 *C*

Where ** TN ** = number of labor-hours required to produce the

*N*th unit

** T 1** = number of labor-hours required to produce the first unit

** C ** = learning-curve coefficient

The learning-curve coefficient, ** C **, depends on both the learning rate and the unit of interest.

Table 3 Learning Curve-Coefficient Approach

Abbildung in dieser Leseprobe nicht enthalten

### 2.3 The Standard time and Normal time

**A. Standard Time**

The standard time is the time required by an average skilled operator, working at a normal pace, to perform a specified task using a prescribed method; it’s also known as the time that a qualified and well trained operator working at a normal pace will need to complete one cycle of the operation.

The standard time includes appropriate allowances to allow the person to recover from fatigue and, where necessary, an additional allowance to cover contingent elements which may occur but have not been observed.

** Standard time = normal time +allowance**

The Standard Time is the product of three factors:

**- Observed time**: The time measured to complete the task.

**- Performance rating factor**: The pace the person is working at; 90% is working slower than normal, 110% is working faster than normal, 100% is normal.

**- Personal**, **Fatigue**, and **Delay** (PFD) allowance.

And the standard time can be determined using the following techniques: Time study, predetermined motion time system (PMTS or PTS), Standard data system, work sampling and Method of calculation. It can also be used in several branches such as : Evaluation of alternative methods, Labor cost control and manpower planning, overhead cost estimation and budgeting, production scheduling: CPM, Production line balancing, Plant layout and plant capacity, Training and performance evaluation, Output-based incentive scheme design .

**[...]**