Chapter 10: Analysis of Cost Behaviour

A complete set of vocabulary flashcards covering cost behaviour, estimation methods, regression, operational time, quality management, and learning curves.

Cost Function

A mathematical description of how a cost changes when the level of an activity changes.

Variable Cost

A cost whose total changes in direct proportion to changes in activity or output (for example, more units produced leading to more materials used).

Fixed Cost

A cost whose total does not change as activity changes; also referred to as a constant.

Mixed Cost

A cost that has both a fixed component and a variable component; also called a semi variable cost.

Linear Cost Function

A mathematical function represented by the equation $Y = a + bx$, where $Y$ is total cost, $a$ is fixed cost, $b$ is variable cost per unit or slope, and $x$ is activity level or cost driver.

Relevant Range

The range of activity where cost behaviour is predictable and approximately linear, outside of which cost behaviour can change.

Cause-and-Effect Relationship

The primary requirement when estimating a cost function, which can arise from a physical relationship, a contractual agreement, or operational knowledge.

Industrial Engineering Method

A cost estimation method that analyzes the physical relationship between inputs and outputs, often using detailed time-and-motion studies, which can be expensive and time-consuming.

Conference Method

A cost estimation method that relies on the subjective opinions, knowledge, and cooperation of individuals across different departments.

Account Analysis Method

A widely used, reasonably accurate, yet subjective cost estimation method that classifies individual accounts as variable, fixed, or mixed.

Quantitative Analysis

An objective cost estimation approach that uses past data and mathematical or statistical methods, such as the high-low method or regression analysis.

High-Low Method

The simplest quantitative cost estimation technique, which determines the cost function using only the highest and lowest activity level observations.

Regression Analysis

A statistical method that measures the relationship between variables by utilizing all available observations to estimate a cost line.

Simple Regression

A regression model that uses exactly one independent variable to estimate costs (such as using machine-hours to predict manufacturing costs).

Multiple Regression

A regression model that uses two or more independent variables to estimate costs (such as using machine-hours and labour-hours to predict manufacturing costs).

Goodness of Fit

A measure showing the strength of the statistical relationship between the cost driver and the cost.

Residual

The difference between the actual observed cost and the estimated cost predicted by the regression line.

Step Costs

Costs that remain constant over small ranges of activity but increase in discrete blocks or lots rather than unit by unit.

Learning Curve

A model showing that as workers gain experience, they become faster and the labour hours required per unit decrease.

Experience Curve

A broader expansion of the learning curve that incorporates efficiency gains across non-manufacturing activities such as marketing and distribution.

Customer-Response Time

The duration of time from when a customer places an order until the finished product or service is delivered.

Manufacturing Cycle/Lead Time

The duration of time from when manufacturing receives an order until the finished product is fully produced.

On-Time Performance

An operational measure tracking whether a product or service is delivered by the scheduled time.

Time Driver

Any factor for which a change causes a direct change in the speed of an activity, primarily including demand uncertainty and bottlenecks.

Bottleneck

An operation or step in a process where the incoming work approaches or exceeds the available capacity.

Prevention Costs

Quality costs incurred to prevent defects and stop problems before they happen.

Appraisal Costs

Quality costs incurred to detect and inspect for defects.

Internal Failure Costs

Quality costs resulting from defective products detected before they are shipped to customers.

External Failure Costs

Quality costs resulting from defective products identified after they have been shipped to customers.

Control Chart

A statistical quality control graph showing observations over time relative to control limits to distinguish between random and nonrandom variation.

Pareto Diagram

A chart that displays defect types ranked from most frequent to least frequent to identify the most common quality problems.

Cause-and-Effect Diagram

Also known as a fishbone diagram; a diagnostic tool used to identify and map potential root causes of quality defects.

Coefficient of Determination ($R^2$)

A regression statistic that measures the percentage of change in the predicted dependent variable explained by changes in the cost driver.

Standard Error

A regression metric that reflects how much an estimated coefficient may be affected by random factors.

$t$-Value ($t$ Stat)

A test statistic used to determine whether a relationship between variables exists that cannot be explained solely by random chance.

Multicollinearity

A condition in multiple regression where two or more predictor variables are highly correlated with each other, typically indicated by a correlation of $\ge 0.70$.

Cumulative Average-Time Learning Model

A learning model where the cumulative average time required per unit decreases by a constant percentage every time the cumulative quantity of units produced doubles.

Incremental Unit-Time Learning Model

A learning model where the time required to produce the last individual unit decreases by a constant percentage every time the cumulative quantity of units produced doubles.

Learning Curve Formula

The mathematical formulation $y = aX^b$, where $y$ is time, $X$ is cumulative units, $a$ is the time required for the first unit, and $b$ is the learning-curve factor calculated as $b = \frac{\ln(\text{Learning Curve } \%)}{\ln(2)}$.