Supply Chain Management CH 5

Demand Forecasting

*Important in pull manufacturing(based on actual demand)* Demand (n)- the state of being wanted or sought for purchase or use.* Forecast (v)- to predict a future condition or occurrence; calculate in advance; (n) a prediction as to something in the future* A forecast is an estimate of future demand & provides the basis for supply planning decisions as well as company financial investment decisions

Qualitative Forecasting Cont'd

1. Jury of executive opinion- An experienced group of senior management executives knowledgeable about the market, competitors, and the business environment develop a forecast 2. Delphi method- Internal and external experts are surveyed about future events and long-term forecasts of demand. Multiple survey rounds until there is a consensus 3.Sales force composite-Utilizes the knowledge of the sales force. The sales force is seen to have recent and accurate information concerning the market and the needs of the customer and are best to make the forecast 4.Consumer survey-Surveys are administered to customers and issues such as future buying habits, new product ideas, and opinions about existing products are gathered and analyzed to determine a forecast.

COLLABORATIVE Planning, Forecasting, & Replenishment (CPFR)

A business practice that combines the intelligence of multiple trading partners in the planning & fulfillment of customer demands. Links sales & marketing best practices, such as category management, to supply chain planning processes to increase availability while reducing inventory, transportation & logistics costs. Real value of CPFR comes from sharing of forecasts among firms rather than sophisticated algorithms from only one firm. Does away with the shifting of inventories among trading partners in the supply chain. CPFR provides many benefits to partners in the supply chain and is best used with Strategic supply chain partners. It requires the keys to successful partnerships be implemented CPFR Model Step 1: Collaboration Arrangement Step 2: Joint Business Plan Step 3: Sales Forecasting Step 4: Order Planning/Forecasting Step 5: Order Generation Step 6: Order Fulfillment Step 7: Exception Management Step 8: Performance Assessment

Components of Time Series forecast

Analyze data to detect the following variations: Trend variations: Long term trend of the demand. Is it increasing or decreasing? Any expected impact to the long term trend (newer technology, change in population, etc) Cyclical variations: wavelike movements that are longer than a year (business cycle). Economic swings (recession, expansion) Seasonal variations: show peaks & valleys that repeat over a consistent interval such as: Hours (breakfast, lunch, dinner) Days (weekdays, weekends, holidays) Weeks (beginning of month, end of month) Months (key activities, holidays) Seasons (holiday, winter, summer, back to school) Random variations: due to unexpected or unpredictable events (Disruptions; storms, strikes, etc)

The Bullwhip Effect

Forecast changes & their corresponding orders along the supply chain can become amplified and accumulate, causing the bullwhip effect whereby excess safety stock is included at all levels of the supply chain from information gaps resulting in higher Supply Chain costs

Forecast Accuracy

Forecast error is the difference between the actual quantity & the forecasted amount. The formula for Forecast error is: et = At - Ft where et = forecast error for Period t At = actual demand for Period t Ft = forecast for Period t Key - need to identify actual error (positive & negative) as well as the absolute error. If actual sales in August was 25,500 units and the forecast for August was 25,000, The forecast error is: +500

Qualitative Forecasting Methods

Generally used when data are limited, unavailable, or not currently relevant. Forecast depends on skill & experience of forecaster(s) & available applicable information. Four qualitative models used are: Jury of executive opinion Delphi method Sales force composite Consumer survey

Demand Forecasting Cont'd

Goal is to minimize forecast error (improve accuracy) Managing demand requires timely & accurate forecasts Factors that influence demand must be considered when forecasting. Accurate forecasting provides: reduced inventories, costs, stock-outs improves production plans & customer service

Simple and Weighted Moving Average

Key Questions: How many periods to use? More periods, less reliance on each. How determine the weights for each period? Simple - all periods have same impact Weighted - can give more emphasis to recent data Balance recent data vs. historical info. **Best to try different methods and compare to actual results to find best fit**

Linear Regression Forecasting Model

Linear Regression Forecasting Model. The trend can be estimated using linear regression to fit a line to a time series of data. Ŷ = b + ax where Ŷ = forecast or dependent variable x = time variable b = intercept of the line a = slope of the line

Measures of Forecasting Accuracy

Mean absolute deviation (MAD)- Identifies how far off the forecast was in absolute terms. Ex. 100 units Mean absolute percentage error (MAPE)- provides a perspective of the true magnitude of the forecast error. Ex. 100 on 1,000 <10%> vs. 100 on 100,000 <1%> Mean squared error (MSE)- large forecast errors are heavily penalized, focuses on minimizing errors Running Sum of Forecast Errors (RSFE) indicates bias in the forecasts or the tendency of a forecast to be consistently higher or lower than actual demand. Tracking signal determines if forecast is within acceptable control limits. If the tracking signal falls outside the pre-set control limits, there is a bias problem with the forecasting method and an evaluation of the way forecasts are generated is warranted. rsfe/MAD=Tracking signal

Examples of When Qualitative Forecast would be used:

New product introduction (new category) Entering a New Market Expansion into new usage

Cause and Effect Forecasting Models

One or several external variables are identified that are related to demand Simple regression- Only one explanatory variable is used & is the same as the linear regression model. **The difference is that the x variable is no longer time but an explanatory variable** Ŷ = b + ax where Ŷ = forecast or dependent variable x = explanatory or independent variable b = intercept of the line a = slope of the line Multiple regression. Several explanatory variables are used to make the forecast. Ŷ = b + a1x1 + a2x2 + . . . akxk where Ŷ = forecast or dependent variable xk(under x) = kth explanatory or independent variable(s) b = intercept of the line ak(under a) = regression coefficient of the independent variable xk Cause & Effect Model One or several external variables are identified that are related to demand: External variable must be a cause of the demand Common error - a variable that follows same trend

Which to Use When

Qualitative forecasting - when historical data is neither available nor relevant. new product introduction, new market, new segment Watch for Bias in the forecast Quantitative forecasting when historical data is available and relevant. Time series - when history is the best indicator Industrial, Retail sales, seasonality, economy Cause & effect when there is clear alignment between cause and effect (not just coincidental) Weather, related product, demographic, event

Forecasting Techniques

Qualitative forecasting is based on using Macro Economic data, opinion & intuition. Quantitative forecasting uses mathematical models & historical data to make forecasts. Time series models are the most frequently used among all the forecasting models. Cause & effect identify causes of a demand and base the forecast on estimation of the cause

Bias may impact the accuracy of Qualitative Forecasting Methods

Since the forecast is based on opinion, it opens the door to Bias. Jury of executive opinion Dominant executive, amount needed to meet ROI commitment Delphi method Surveyor can influence the outcome Sales force composite Sales commissions impact accuracy of information provided Consumer survey Consumer actions don't always match their words

Quantitative Forecasting Methods

Time series forecasting- based on the assumption that the future is an extension of the past. Historical data is used to predict future demand. Cause & Effect forecasting- assumes that one or more factors (independent variables) predict future demand.

Simple Moving Average Forecasting Model

Uses historical data to generate a forecast. Works well when demand is stable over time.

Weighted Moving Average Forecasting Models

based on an n-period weighted moving average. Works well when demand is stable over time with an increasing or decreasing trend

Naiive Forecast

the estimate of the next period is equal to the demand in the past period. Ft+1 = At Where Ft+1 = forecast for period t+1 At = actual demand for period t If actual sales in August was 25,500 units, the forecast for September will be 25,500.