SCM Forecasting  SCRC Articles Library
SCM Approaches to Forecasting : A Tutorial Articles

Demand Forecasting: A Cross Discipline Perspective Part II: Probabilistic and Ensemble Forecasting
In forecasting for demand there are two predominant methods of modeling available: deterministic and probabilistic. Normally just saying the names of these techniques is enough to turn someone off from the topic, however, it is important to understand the strengths and weaknesses of each. Deterministic is simply defined as a forecast in which the results of the model are completely determined by present conditions (Lewis 2005). Simply stated, forecasted demand is completely and solely dependent on what we know right here and now. This sounds somewhat absurd since we know market volatility and global economic conditions can change the demand outlook

Demand Forecasting: A Cross Discipline Perspective Part I: Model Output Statistics (MOS)
This paper will highlight some of the numerical modeling techniques used in weather forecasting that can be applied to improve business forecasting. Demand and weather forecasting are comparable in so much as they are both complex environments in which many variables influence the outcome. The weather community has developed many advanced modeling techniques to provide more accurate forecasts, however, it has been the advancement of computing power and technology that has held back many of these techniques from becoming operational. With the need to improve forecasts while waiting for technology to improve, several modeling methods have been developed which produce

References: Approaches to Forecasting : A Tutorial
References Most introductory textbooks in Operations and Supply Chain Management offer a good discussion of forecasting models. Techniques such as time series models, regression, and measures of forecasting accuracy are routinely covered in these books. One possible source: Cecil Bozarth and Robert Handfield, Introduction to Operations and Supply Chain Management, PrenticeHall, 2006. ISBN 0139446206 Planners who are interested in a more advanced treatment of forecasting might want to check devoted to business forecasting. One possible source: John Hanke, Arthur Reitsch, and Dean Wichern, Business Forecasting, PrenticeHall, 2001. ISBN 0130878103.
Categories:

Summary: Approaches to Forecasting : A Tutorial
Summary Forecasting is a critical part of CPFR. Fortunately, there are a number of welldeveloped tools and strategies for developing forecasts. Avoid forecasting values when you can calculate them. For example, in an ideal world, we would only forecast POS demand, and use these numbers to calculate the replenishment forecasts at the DCs, and the order forecasts at the plants. Quantitative forecasting techniques are bestsuited to situations where historical data exists and the past is considered a good indicator of what will happen in the future. Qualitative forecasting methods are used when the situation is vague and little data exists
Categories:

Q2 Forecasting: Approaches to Forecasting : A Tutorial
Q2 Forecasting What Is Q2 Forecasting? QuantitativeQualitative Forecasting Basic ideas behind Q2 forecasting: Wherever possible, forecasters should first develop a quantitative forecast, then use this for the basis for more qualitative analysis. Historical data is never a perfect indicator of what will happen in the future, but this doesn’t excuse us from considering it! By following the Q2 principle, forecasters can: Keep their intuition grounded in the data. Expose the assumptions behind qualitative forecasts (ex. – “Why do we expect demand to be three times higher than last year?”).
Categories:

3  Step Process: Approaches to Forecasting : A Tutorial
3 – Step Process STEP 1: Divide Items into Three Major Categories STEP 2: Apply the simplest tools needed to do the job STEP 3: Spend effort on the “difficult few” Other Considerations Q2 Forecasting h2. STEP 1: Divide Items into Three Major Categories STEP 2: Apply the simplest tools needed to do the job STEP 3: Spend effort on the “difficult few” Other Considerations Begin to develop the database needed to support quantitative analysis: Sales and order history Promotions, price discounts offered, etc. Measure forecast accuracy and put in place a formal review step to review what forecasting methods did / did not work Reassign items to categories /
Categories:

Qualitative Methods :Measuring Forecast Accuracy : A Tutorial
Qualitative Methods Common Qualitative Forecasting Methods EXAMPLE: Life Cycle analogy Analyzing the Life Cycle Data for the Previous Version Questions to Consider When Using the Life Cycle Analogy to Forecast for a New Product Common Qualitative Forecasting Methods Executive and outsider opinions Sales force composite This involves having product managers or sales reps developing individual forecasts, and then adding them up Panel consensus & Delphi method Both methods have experts work together to develop forecasts The Delphi method has experts develop forecasts individually, then share their findings. The process is repeated until a consensus emerges. Life cycle analogy Used when the
Categories:

Multiple Regression: Approaches to Forecasting : A Tutorial
Multiple Regression Advanced techniques can be used when there is trend or seasonality, or when other factors (such as price discounts) must be considered. What is Multiple Regression? Resulting Forecast Model Comparing Multiple Regression Model Results against Historic Demand h2. What is Multiple Regression? Analogous to single regression, but allows us to have multiple predictor variables: Y = a + b1*X1 + b2*X2 + b3*X3 … *Practically speaking, there is a limit to the number of predictor variables you can have without violating some statistical rules. In most cases, 2 or 3 predictor variables should be plenty. In this case, we have 24 months of
Categories:

Measuring Forecast Accuracy: Approaches to Forecasting : A Tutorial
Measuring Forecast Accuracy How Do We Measure Forecast Accuracy? Measures of Forecast Accuracy Mean Forecast Error (MFE) Mean Absolute Deviation (MAD) Tracking Signal Other Measures How Do We Measure Forecast Accuracy? Used to measure: Forecast model bias Absolute size of the forecast errors Can be used to: Compare alternative forecasting models Identify forecast models that need adjustment (management by exception) Measures of Forecast Accuracy Error = Actual demand – Forecast OR et = At – Ft h2. Mean Forecast Error (MFE) For n time periods where we have actual demand and forecast values: Ideal value = 0; MFE > 0, model tends to underforecast MFE < 0, model
Categories:

Single Regression: Approaches to Forecasting : A Tutorial
Single Regression Advanced techniques can be used when there is trend or seasonality, or when other factors (such as price discounts) must be considered. What is Single Regression? EXAMPLE: 16 Months of Demand History EXAMPLE: Building a Regression Model to Handle Trend and Seasonality EXAMPLE: Causal Modeling h2. What is Single Regression? Develops a line equation y = a + b(x) that best fits a set of historical data points (x,y) Ideal for picking up trends in time series data Once the line is developed, x values can be plugged in to predict y (usually demand) For time series models, x
Categories:

Double Exponential Smoothing: Approaches to Forecasting : A Tutorial
Double Exponential Smoothing What Is Double Exponential Smoothing? Time Series with Trend: Double Exponential Smoothing h2. What Is Double Exponential Smoothing? …like regular exponential smoothing, except includes a component to pick up trends. Time Series with Trend: Double Exponential Smoothing Formula Ft = Unadjusted forecast (before trend) Tt = Estimated trend AFt = Trendadjusted forecast Ft = a* At1 + (1 a) * (Ft1 + Tt1) Tt = b* (At1Ft1) + (1 b) * Tt1 AFt = Ft + Tt To start, we assume no trend and set our “initial” forecast to Period 1 demand. … We then calculate our forecast for Period 2. … But Period 2 demand turns out to be 20. What is the
Categories:

Exponential Smoothing: Approaches to Forecasting : A Tutorial
Exponential Smoothing What is Exponential Smoothing? Exponential Smoothing Forecaset with a = .3 h2. What is Exponential Smoothing? A type of weighted moving averaging model Part of many forecasting packages; ideal for developing forecasts of lots of smaller items Needs only three numbers: Ft1 = Forecast for the period before current time period t At1 = Actual demand for the period before current time period t a = Weight between 0 and 1 Formula As a gets closer to 1, the more weight put on the most recent demand number h2. Exponential Smoothing Forecaset with a = .3
Categories:

Time Series Models: Approaches to Forecasting : A Tutorial
Time Series Models Time Series Models Time Series Components of Demand… Basic Idea Behind Time Series Models Moving Average Models Table of Forecasts & Demand Values… ... and Resulting Graph What Are Time Series Models Quantitative forecasting models that use chronologically arranged data to develop forecasts. Assume that what happened in the past is a good starting point for predicting what will happen in the future. These models can be designed to account for: Randomness Trend Seasonality effects Advantages Can quickly be applied to a large number of products Forecast accuracy measures can be used to identify forecasts that need adjustment (management
Categories:

Basic Rules of Forecasting: Approaches to Forecasting : A Tutorial
Basic Rules of Forecasting p. Forecasts Forecasts Are No Substitue for Calculated Values Two Distinct Approaches to Forecasting h2. What Are the Basic Rules of Forecasts Forecasts are almost always wrong. Important to measure forecast accuracy and take actions to improve when necessary Nearterm forecasts tend to be more accurate. Forecasts for groups (product categories, multiple stores, etc.) tend to be more accurate. Forecasts are no substitute for calculated values. h2. Forecasts Are No Substitue for Calculated Values Recall our earlier example… POS forecasts were used to calculate the replenishment forecast at the DC Disconnect h2. Two Distinct Approaches to
Categories:

Introduction: Approaches to Forecasting : A Tutorial
Introduction p. Learning Objectives Requirements of Forecasting Section Outline Learning Objectives By the end of this module, you will be able to: List the basic rules of forecasting, and explain what is meant by the rule, “Forecasts are no substitute for calculated demand.” Develop and interpret simple time series forecasting models. Develop and interpret simple and multiple regression forecasting models, and use regression to develop both time series and causal forecasts models. Calculate and interpret measures of forecast accuracy. Explain when quantitative versus qualitative forecasting techniques should be used, and the advantages and disadvantages of each. Develop forecasting strategies for
Categories: