Demand Forecasting Fall, 2016 EMBA 512 Demand Forecasting

Slides:



Advertisements
Similar presentations
Forecasting OPS 370.
Advertisements

Operations Management Forecasting Chapter 4
Operations Management For Competitive Advantage © The McGraw-Hill Companies, Inc., 2001 C HASE A QUILANO J ACOBS ninth edition 1Forecasting Operations.
Time Series Analysis Autocorrelation Naive & Simple Averaging
T T18-03 Exponential Smoothing Forecast Purpose Allows the analyst to create and analyze the "Exponential Smoothing Average" forecast. The MAD.
Qualitative Forecasting Methods
Chapter 12 - Forecasting Forecasting is important in the business decision-making process in which a current choice or decision has future implications:
Forecasting.
Operations Management R. Dan Reid & Nada R. Sanders
Operations Management Forecasting Chapter 4
© 2004 by Prentice Hall, Inc., Upper Saddle River, N.J Operations Management Forecasting Chapter 4.
T T18-06 Seasonal Relatives Purpose Allows the analyst to create and analyze the "Seasonal Relatives" for a time series. A graphical display of.
Slides 13b: Time-Series Models; Measuring Forecast Error
Fall, 2012 EMBA 512 Demand Forecasting Boise State University 1 Demand Forecasting.
Forecasting Chapter 15.
Forecasting.
Chapter 15 Demand Management & Forecasting
The Importance of Forecasting in POM
Demand Management and Forecasting
Planning Demand and Supply in a Supply Chain
3-1 McGraw-Hill/Irwin Operations Management, Seventh Edition, by William J. Stevenson Copyright © 2002 by The McGraw-Hill Companies, Inc. All rights reserved.
Operations Management
3-1Forecasting William J. Stevenson Operations Management 8 th edition.
3-1Forecasting. 3-2Forecasting FORECAST:  A statement about the future value of a variable of interest such as demand.  Forecasts affect decisions and.
© 2004 Prentice-Hall, Inc. Chapter 7 Demand Forecasting in a Supply Chain Supply Chain Management (2nd Edition) 7-1.
1 DSCI 3023 Forecasting Plays an important role in many industries –marketing –financial planning –production control Forecasts are not to be thought of.
DAVIS AQUILANO CHASE PowerPoint Presentation by Charlie Cook F O U R T H E D I T I O N Forecasting © The McGraw-Hill Companies, Inc., 2003 chapter 9.
Operations Management For Competitive Advantage 1Forecasting Operations Management For Competitive Advantage Chapter 11.
MBA.782.ForecastingCAJ Demand Management Qualitative Methods of Forecasting Quantitative Methods of Forecasting Causal Relationship Forecasting Focus.
Introduction to Forecasting IDS 605 Spring Forecasting 4 A forecast is an estimate of future demand.
1-1 1 McGraw-Hill/Irwin ©2009 The McGraw-Hill Companies, All Rights Reserved.
Forecasting February 26, Laws of Forecasting Three Laws of Forecasting –Forecasts are always wrong! –Detailed forecasts are worse than aggregate.
To Accompany Ritzman & Krajewski, Foundations of Operations Management © 2003 Prentice-Hall, Inc. All rights reserved. Chapter 9 Demand Forecasting.
Forecasting Operations Management For Competitive Advantage.
Demand Management and Forecasting Module IV. Two Approaches in Demand Management Active approach to influence demand Passive approach to respond to changing.
Operations Fall 2015 Bruce Duggan Providence University College.
Forecasting. 預測 (Forecasting) A Basis of Forecasting In business, forecasts are the basis for budgeting and planning for capacity, sales, production and.
Reid & Sanders, Operations Management © Wiley 2002 Forecasting 8 C H A P T E R.
1 Chapter 13 Forecasting  Demand Management  Qualitative Forecasting Methods  Simple & Weighted Moving Average Forecasts  Exponential Smoothing  Simple.
McGraw-Hill/Irwin © 2006 The McGraw-Hill Companies, Inc., All Rights Reserved. 1.
McGraw-Hill/Irwin Copyright © 2008 by The McGraw-Hill Companies, Inc. All rights reserved. Demand Management and Forecasting CHAPTER 10.
CHAPTER 12 FORECASTING. THE CONCEPTS A prediction of future events used for planning purpose Supply chain success, resources planning, scheduling, capacity.
3-1Forecasting CHAPTER 3 Forecasting McGraw-Hill/Irwin Operations Management, Eighth Edition, by William J. Stevenson Copyright © 2005 by The McGraw-Hill.
Forecasting Demand. Problems with Forecasts Forecasts are Usually Wrong. Every Forecast Should Include an Estimate of Error. Forecasts are More Accurate.
3-1Forecasting William J. Stevenson Operations Management 8 th edition.
13 – 1 Copyright © 2010 Pearson Education, Inc. Publishing as Prentice Hall. Forecasting 13 For Operations Management, 9e by Krajewski/Ritzman/Malhotra.
McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All Rights Reserved. Chapter 3 Forecasting.
Demand Management and Forecasting Chapter 11 Portions Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin.
Forecasting. ©2006 Pearson Prentice Hall — Introduction to Operations and Supply Chain Management — Bozarth & Handfield Chapter 9, Slide 2 Why Forecast?
Chapter 11 – With Woodruff Modications Demand Management and Forecasting Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved.McGraw-Hill/Irwin.
T T18-02 Weighted Moving Average Forecast Purpose Allows the analyst to create and analyze the "Weighted Moving Average" forecast for up to 5.
Welcome to MM305 Unit 5 Seminar Dr. Bob Forecasting.
Welcome to MM305 Unit 5 Seminar Forecasting. What is forecasting? An attempt to predict the future using data. Generally an 8-step process 1.Why are you.
Chapter 7 Demand Forecasting in a Supply Chain Supply Chain Management 7-1.
Demand Forecasting: Time Series Models Professor Stephen R
Forecasting Chapter 9.
Forecasts.
Operations Management Contemporary Concepts and Cases
Forecasting Methods Dr. T. T. Kachwala.
Chapter 7 Demand Forecasting in a Supply Chain
Demand Management and Forecasting
Forecasting Chapter 11.
Fall, 2017 EMBA 512 Demand Forecasting
FORCASTING AND DEMAND PLANNING
McGraw-Hill/Irwin Copyright © 2009 by The McGraw-Hill Companies, Inc. All rights reserved.
Chapter 7 Demand Forecasting in a Supply Chain
Principles of Supply Chain Management: A Balanced Approach
Forecasting is an Integral Part of Business Planning
Demand Management and Forecasting
Forecasting Plays an important role in many industries
Presentation transcript:

Demand Forecasting Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Objectives Understand the role of forecasting Understand the issues Understand basic tools and techniques Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecasting Developing predictions or estimates of future values Demand volume Price levels Lead times Resource availability ... Fall, 2016 EMBA 512 Demand Forecasting Boise State University

The Role of Forecasting Necessary Input to all Planning Decisions Operations: Inventory, Production Planning & Scheduling Finance: Plant Investment & Budgeting Marketing: Sales-Force Allocation, Pricing Promotions Human Resources: Workforce Planning Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Demand Forecasting For manufactured items and conventional goods, forecasts are used to determine Replenishment levels and safety stocks Set production plans Determine procurement schedules Capacity planning, financial planning, & workforce planning Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Demand Forecasting For services, demand forecasts are used for Capacity planning, workforce scheduling, procurement & budgeting. Because services cannot be stored, demand forecasting for services is often concerned with forecasting the peak demand, rather than the average demand and its range. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Characteristics of Forecasts Forecast are always wrong. A good forecast is more than a single value. Forecast accuracy decreases with the forecast horizon. Aggregate forecasts are more accurate than disaggregated forecasts. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Independent vs. Dependent Demand Exogenously controlled Subject to random or unpredictable changes What we forecast Dependent or Derived Calculated or derived from other sources Do not forecast Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecasting Methods Qualitative or Judgmental Ask people who ought to know Historical Projection or Extrapolation Time Series Models Moving Averages Exponential Smoothing Regression based methods Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Basic Approach to Demand Forecasting Identify the Objective of the Forecast Integrate Forecasting with Planning Identify the Factors that Influence the Demand Forecast Identify the Appropriate Forecasting Model Monitor the Forecast (Measure Errors) Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series Methods Appropriate when future demand is expected to follow past demand patterns. Future demand is assumed to be influenced by the current demand, as well as historical growth and seasonal patterns. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series Models With time series models observed demand can be broken down into two components: systematic and random. Observed Demand = Systematic Component + Random Component Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series Methods The systematic component is the expected demand value. It is comprised of the underlying average demand, the trend in demand, and the seasonal fluctuations (seasonality) in demand. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Idea Behind Time Series Models Distinguish between random fluctuations and true changes in underlying demand patterns. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series Components of Demand Random component Time Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Monthly chart of the DJIA's changes from month to month along with a 3 period simple moving average. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series Methods The random component cannot be predicted. However, its size and variability can be estimated to provide a measure of forecast error. The objective of forecasting is to filter the random component and model (estimate) the systematic component. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Moving Averages Simple, widely used Reduce random noise One Extreme Prediction next period = Demand this period Another Extreme Prediction next period = Long run average Intermediate View Prediction next period = Average of last n periods Fall, 2016 EMBA 512 Demand Forecasting Boise State University

 Moving Average Models Period Demand 1 12 2 15 3 11 4 9 5 10 6 8 7 14 1 12 2 15 3 11 4 9 5 10 6 8 7 14 8 12  3-period moving average forecast for Period 8: = (14 + 8 + 10) / 3 = 10.67 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Weighted Moving Averages Forecast for Period 8 = [(0.5  14) + (0.3  8) + (0.2  10)] / (0.5 + 0.3 + 0.2) = 11.4 What are the advantages? What do the weights add up to? Could we use different weights? Compare with a simple 3-period moving average. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Table of Forecasts and Demand Values . . . Period Actual Demand Two-Period Moving Average Forecast Three-Period Weighted Moving Average Forecast Weights = 0.5, 0.3, 0.2 1 12   2 15 3 11 13.5 4 9 13 12.4 5 10 10.8 6 8 9.5 9.9 7 14 8.8 11.4 11.8 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

. . . and Resulting Graph Note how the forecasts smooth out variations Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Simple Exponential Smoothing Sophisticated weighted averaging model Needs only three numbers: Ft = Forecast for the current period t Dt = Actual demand for the current period t a = Weight between 0 and 1 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Exponential Smoothing Moving Averages Equal weight to older observations Exponential Smoothing More weight to more recent observations Forecast for next period is a weighted average of Observation for this period Forecast for this period Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Simple Exponential Smoothing Formula Ft+1 = Ft + a (Dt – Ft) = a × Dt + (1 – a) × Ft Where did the current forecast come from? What happens as a gets closer to 0 or 1? Where does the very first forecast come from? Very first forecast is often set equal to the actual demand to start the process. An alternate approach is to set the first forecast to the moving average of the previous two or three months. Alpha should be large if the demand data is relatively stable, small if the demand data varies quite a bit. Otherwise it takes a long time for the forecast to converge on relatively smooth demand (overdamped correction) and the forecast overshoots the variations for fluctuating demand (underdamped correction) Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Exponential Smoothing Forecast with a = 0.3 Period Actual Demand Exponential Smoothing Forecast 1 12 11.00 (given) 2 15 11.30 3 11 12.41 4 9 11.99 5 10 11.09 6 8 10.76 7 14 9.93 11.15   11.41 F2 = 0.3×12 + 0.7×11 = 3.6 + 7.7 = 11.3 F3 = 0.3×15 + 0.7×11.3 = 12.41 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Resulting Graph Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time Series with random and trend components Demand Time Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Linear Trend Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Exponential Trend Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Trends What do you think will happen to a moving average or exponential smoothing model when there is a trend in the data? Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Simple Exponential Smoothing Always Lags A Trend Period Actual Demand Exponential Smoothing Forecast 1 11 11.00 2 12 3 13 11.30 4 14 11.81 5 15 12.47 6 16 13.23 7 17 14.06 8 18 14.94 9   15.86 Because the model is based on historical demand, it always lags the obvious upward trend Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Simple Linear Regression Time Series Find best fit of proposed model to past data Project that fit forward Assumes a linear relationship: y = a + b(x) y x Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Definitions Y = a + b(X) Y = predicted variable (i.e., demand) X = predictor variable “X” is the time period for linear trend models. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Example: Regression Used to Estimate A Linear Trend Line Period (X) Demand (Y) 1 110 2 190 3 320 4 410 5 490 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Resulting Regression Model: Forecast = 10 + 98×Period Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Time series with random, trend and seasonal components Demand June Class discussion: what could account for this? Lawnmower sales? Camping trailer sales? Vacation package sales? June June June June Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Trend & Seasonality Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Seasonality Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Modeling Trend & Seasonal Components Quarter Period Demand Winter 07 1 80 Spring 2 240 Summer 3 300 Fall 4 440 Winter 08 5 400 Spring 6 720 Summer 7 700 Fall 8 880 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

What Do You Notice? Forecasted Demand = –18.57 + 108.57 x Period Actual Demand Regression Forecast Forecast Error Winter 07 1 80 90 -10 Spring 2 240 198.6 41.4 Summer 3 300 307.1 -7.1 Fall 4 440 415.7 24.3 Winter 08 5 400 524.3 -124.3 6 720 632.9 87.2 7 700 741.4 -41.4 8 880 850 30 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Regression picks up trend, but not the seasonality effect Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Calculating Seasonal Index: Winter Quarter (Actual / Forecast) for Winter Quarters: Winter ‘07: (80 / 90) = 0.89 Winter ‘08: (400 / 524.3) = 0.76 Average of these two = 0.83 Interpret! The normal trend line prediction needs to be adjusted downward for Winter quarters. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Seasonally Adjusted Forecast Model For Winter Quarter [ –18.57 + 108.57×Period ] × 0.83 Or more generally: [ –18.57 + 108.57 × Period ] × Seasonal Index Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Seasonally Adjusted Forecasts Forecasted Demand = –18.57 + 108.57 x Period Period Actual Demand Regression Forecast Demand/Forecast Seasonal Index Seasonally Adjusted Forecast Forecast Error Winter 07 1 80 90 0.89 0.83 74.33 5.67 Spring 2 240 198.6 1.21 1.17 232.97 7.03 Summer 3 300 307.1 0.98 0.96 294.98 5.02 Fall 4 440 415.7 1.06 1.05 435.19 4.81 Winter 08 5 400 524.3 0.76 433.02 -33.02 6 720 632.9 1.14 742.42 -22.42 7 700 741.4 0.94 712.13 -12.13 8 880 850 1.04 889.84 -9.84 Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Would You Expect the Forecast Model to Perform This Well With Future Data? Fall, 2016 EMBA 512 Demand Forecasting Boise State University

The Perfect (Imaginary) Forecast Fall, 2016 EMBA 512 Demand Forecasting Boise State University

A More Realistic Forecast Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecast Error Building a Forecast Forecast Error Fit to historical data Project future data Forecast Error How well does model fit historical data Do we need to tune or refine the model Can we offer confidence intervals about our predictions Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecast Error The forecast error measures the difference between the actual demand and the forecast of demand. The forecast is based on the systematic component and the random component is estimated based on the forecast error. Forecast Error = Actual – Forecast Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Measures of Forecast Accuracy Forecast Errort (Et)= Demandt-Forecastt Mean Squared Error (MSE) Mean Absolute Deviation (MAD) Bias Tracking Signal Relative Forecast Errors Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Mean Squared Error (MSE) The MSE estimates the variance of the forecast error. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Mean Absolute Deviation (MAD) The MAD can be used to estimate the standard deviation of the random component, assuming the random component is normally distributed: σ = 1.25MAD Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Bias To determine whether a forecasting method consistently over-or- underestimates demand, calculate the sum of the forecast errors: Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Tracking Signal The tracking signal (TS) is the ratio of the bias to the MAD. Tracking signals outside the range + 6 indicates that the forecast is biased and either under predicting (negative) or over predicting (positive) demand. Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecast Accuracy & Demand Variability (Normally Distributed Demand) Coefficient of Variation Probability Demand is Within 25% of the Forecast 0.10 98.76% 0.25 68.27% 0.50 38.29% 0.75 26.11% 1.00 19.74% 1.50 13.24% 2.00 9.95% 3.00 6.64% Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Issues Forecasting is a necessary evil, try to reduce the need for it. Complexity costs money, does it provide better forecasts? Aggregation provides accuracy, but precludes local information Forecast the right thing Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecasting Success Story Taco Bell Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Taco Bell Labor is 30% of revenue Make to order environment Feed the dog Taco Bell Labor is 30% of revenue Make to order environment Significant “seasonality” 52% of days sales during lunch 25% of days sales during busiest hour Balance staff with demand Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Value Meals Drove demand Forecasting system in each store forecasts arrivals within 15 minute intervals Simulation system “predicts” congestion and lost sales Optimization system Finds the minimum cost allocation of workers Fall, 2016 EMBA 512 Demand Forecasting Boise State University

Forecasting System Customer arrivals by 15-minute interval of day (e.g., 11:15-11:30 am Friday) Fed by in-store computer system 6-week moving average Estimated savings: Over $40 Million in 3 years. Fall, 2016 EMBA 512 Demand Forecasting Boise State University