Supply Chain Ch. 5

Time Series Forecasting Models

1. Naive Forecast 2. Simple Moving Average 3. Weighted Moving Average 4. Exponential Smoothing

Collaborative Planning, Forecasting, and Replenishment (CPFR)

Aims to enhance supply chain integration by supporting and assisting joint practices. CPFR seeks cooperative management of inventory through joint visibility and replenishment of products throughout the supply chain. Information shared between suppliers and retailers aids in planning and satisfying customer demands through a supportive system of shared information. This allows for continuous updating of inventory and upcoming requirements, essentially making the end-to-end supply chain process more efficient. Efficiency is also created through the decrease of expenditures for merchandising, inventory, logistics, and transportation across all trading partners.

Cause and Effect Forecasting

Assumes that one or more factors are related to demand and, therefore, can be used to predict future demand. Models: 1. Simple Linear Regression Forecast 2. Multiple Regression Forecast

Forecast Bias

Measures the tendency of a forecast to be consistently higher or lower than actual demand over time. Positive

Forecast Error

The difference between the actual quantity and the forecast Expressed as: et= At-Ft

Naive Forecasting

The estimate for the next period is equal to the actual demand for the immediate past period

Weighted Moving Average Forecast

The weighted average of the n-period observations, using unequal weights. Weights should be nonnegative and equal to 1

Quantitative Forecasting Methods

Use mathematical models and and relevant historical data to generate forecasts; Divided into 2 groups: Time series and associative models; All quantitative forecasting models become less accurate as the forecast's time horizon increases

Tracking Signal

Used to determine if the forecast bias is within the acceptable control limits. Expressed as: RSFE/MAD If the tracking signal falls outside preset control limits, there is a bias problem with the forecasting method

Multiple Regression Forecast

Used when there are several explanatory variables to predict the dependent variable. Works well when the relationships between demand and several other factors impacting demand are strong and stable over time

Simple Moving Average Forecast

Uses historical data to generate a forecast and works well when the demand is fairly stable over time

Running Sum of Forecast Errors

An indicator of bias in the forecasts. Positive RSFE indicates that the forecasts are generally lower than actual demand which can lead to stockouts. Negative RSFE shows shows that the forecasts are generally higher than actual demand which can result in excess inventory carrying costs

Mean Absolute Deviation

An indicator of forecast accuracy based on an average of the absolute value of the forecast errors over a given period of time. The measure indicates, on average, how many units the forecast is off from the actual data

Mean Absolute Percentage Error

An indicator of forecast accuracy based on the true magnitude of the forecast error. The monthly absolute forecast error divided by actual demand is summed, then divided by the number of months used in the forecast to derive and average, and lastly multiplied by 100. The measure indicates, on average, what percent the forecast is off from the actual data

Mean Square Error

An indicator of forecast accuracy. The forecast errors are squared and then summed and divided by the number of periods to determine the mean square error. The measure penalizes large errors more than small errors

Cloud-Based Forecasting

Using supplier-hosted or software-as-a-service advanced forecasting applications that are provided to companies on a subscription basis

Simple Linear Regression Forecast

When there is only one explanatory variable, we have a simple regression model equivalent to the linear trend model described earlier. However the x variable in this model is not time but rather an explanatory variable of demand

Customer Surveys

A forecasting questionnaire can be developed that uses input from customers on important issues such as future purchasing needs, new product ideas, and opinions about existing or new products. The data collected is then analyzed and forecasts are developed from the results

Delphi Method

A group of internal and external experts are surveyed during several rounds in terms of future events and long term forecasts of demand, in hopes of converging on a consensus forecast. The summary of responses is then sent out to all of the experts in the next round wherein the individual experts can modify their responses based on the group's response summary. Process continues until a consensus is reached. Applicable for high-risk technology forecasting, large expensive projects, or major new product introductions

Jury of Executive Opinion

A group of the firm's senior management executives who are knowledgeable about their markets, their competitors, and the business environment collectively develop the forecast.; Applicable for long range planning and new product introductions, but is also commonly used for general demand forecasting

Exponential Smoothing

A sophisticated weighted moving average forecasting technique in which the forecast for the next period's demand is the current period's forecast adjusted by a fraction of the difference between the current period's actual demand and forecast. Only two data points are needed and this model is suitable for data that shows little trend or seasonal patterns

Qualitative Forecasting Methods

Based on opinions and intuition and are generally used when data is limited, unavailable, or not currently relevant; 4 common qualitative forecasting models: Jury of executive opinion, delphi method, sales force composite, customer surveys

Time Series Forecasting Components

Based on the assumption that the future is an extension of the past, thus, historical data can be used to predict future demand. Components of a Time Series Forecast: 1. Trend Variations: Represent either increasing or decreasing movements over many years; Common trend lines are linear, s curve, exponential, or asymptotic 2. Cyclical Variations: Wavelike movements that are longer than a year and are influenced by macroeconomic and political factors 3. Seasonal Variations: Show peaks and valleys that repeat over a consistent interval such as hours, days, weeks, months, years, or seasons 4. Random Variations: Due to unexpected or unpredictable events such as natural disasters, strikes, and wars

Linear Trend Forecast

Can be estimated used simple linear regression to fit a line to a series of data occurring over time. The trend line is determined using the least squares method, which minimizes the sum of the squared deviations to determine the characteristics of the linear equation

Sales Force Composite

Generated based on the sales force's knowledge of the market and estimates of customer needs.