Chapter 5 Demand Forecasting TRUE/FALSE 1. One of the goals of an effective CPFR system is to minimize the negative impacts of the bullwhip effect on supply chains. 2. The modern day business environment must deal with a more homogenous consumer base, which has caused the evolution of a more "push" oriented environment where suppliers must focus on manufacturing high volumes of standardized goods. 3. Some of the benefits of CPFR include strengthening partner relationships, providing an analysis of sales and order forecasts both upstream and downstream, and allowing collaboration on future requirements and planning. 4. The true value of CPFR comes from the sophisticated forecasting algorithms that provide companies with highly accurate forecasts, not from the exchange of forecasting information. 5. The difference between a simple regression forecast and a multiple regression forecast is that simple regression is used when there is only one explanatory (or independent) variable, while multiple regression is used when there are numerous explanatory variables. 6. Some of the important steps involved in implementation of a CPRF process model include developing a collaborative arrangement, creating a sales forecast, creating an order forecast, and generating those orders. 7. Some of the leading suppliers of CPFR solutions mentioned in the textbook are JDA Software Group, i2 Technologies, and Oracle. 8. Quantitative forecasting methods are based on opinions and intuition, whereas qualitative forecasting methods use mathematical models and relevant historical data to generate forecasts. 9. If you were calculating a forecast using an exponential smoothing model, a calculation using α = 0.2 would be putting a greater emphasis on recent data, while a calculation using α = 0.8 would be putting a greater emphasis on past data. Thus a lower α is more responsive to changes in demand in the most recent periods.
10. If you felt that recent demand trends were more significant, and thus should be emphasized more in formulating a forecast, then in forecasting demand for the upcoming demand period, you would probably favor using a simple moving average over the conventional weighted moving average. 11. As tighter control limits are instituted for the tracking signal, there is a greater probability of finding exceptions that require no action, but it also means catching changes in demand earlier. 12. In the Delphi forecasting method, a group of internal and external experts are surveyed during several rounds in terms of future events and long-term forecasts of demand but the group members do not physically meet. 13. The Institute of Supply Management (ISM) surveys more than 300 purchasing and supply executives in the United States using a questionnaire seeking information on "changes in production, new orders, new export orders, imports, employment, inventories, prices, lead-times, and the timeliness of supplier deliveries in their companies comparing the current month to the previous month." The ISM Report on Business focuses only on the manufacturing sector. 14. Examples of forecasting accuracy measures are Mean Absolute Deviation, Mean Absolute Percentage Error, and Mean Square Error 15. The objective of CPFR is to optimize the supply chain by improving demand forecast accuracy, delivering the right product at the right time to the right location, reducing inventories across the supply chain, avoiding stock-outs, and improving customer service. As a result of the many benefits attributed to CPFR, this business practice has been widely adopted. MULTIPLE CHOICE 1. What does the acronym CPFR represent? a. Coordinated planning and forecasting relationships b. Collaborative planning, forecasting, and replenishment c. Centralized purchasing and forecasting relationships d. Collaborative purchasing, forecasting, and receivables B PTS: 1 2. Which of the following indices provided by the Institute for Supply Management (ISM) is considered the most important by economists because it is a composite of five weighted, seasonally adjusted indices?
a. Purchasing Managers Index b. Export Orders Index c. Production and Inventory Index d. New Orders Index A PTS: 1 3. According to the textbook, which of the following is NOT a way to closely match supply and demand? a. Holding high amounts of inventory b. Maintaining a rigid pricing system c. Utilizing overtime d. Hiring temporary workers B PTS: 1 4. According to textbook, the top three challenges for CPFR implementation include all of the following except: a. Making organizational and procedural changes b. Trust between supply chain partners c. Cost d. Supplier lead times D PTS: 1 5. Some measures of forecasting accuracy include mean absolute deviation, mean absolute percentage error, and mean squared error. The formula for each is dependent on the forecast error, which is calculated by using the equation: a. Actual demand for period t divided by the forecasted demand for period t b. Actual demand for period t plus the forecasted demand for period t c. Actual demand for period t minus the forecasted demand for period t d. The average of Actual demand for period t and forecasted demand for period t C PTS: 1 6. If a tracking signal is positive, which one of the following is true? a. Actual value is higher than forecast b. Actual value is less than forecast c. Actual value is equal to forecast d. Unable to draw any conclusion A PTS: 1 7. The exponential smoothing forecast has the same value as the naïve forecast when α in the exponential smoothing model is equal to: a. 0 b. 0.5 c. 1 d. Insufficient information provided to determine answer C PTS: 1 Data Set E1 Period Sales Volume 1 10000 2 12400
3 14250 4 15750 5 20500 6 18500 7 15750 8 20500 9 21500 10 22550 8. Using Data Set E1, what would be the forecast for period 7 using a four period moving average: (Choose the closest answer.) a. 17625 b. 15225 c. 15300 d. 17250 D PTS: 1 9. Using Data Set E1, what would be the forecast for period 6 using a five period weighted moving average? The weights for each period are 0.05, 0.10, 0.20, 0.30, and 0.35 from the oldest period to the most recent period, respectively. (Choose the closest answer.) a. 16500 b. 17825 c. 14575 d. 16275 A PTS: 1 10. Using Data Set E1, what would be the forecast for period 6 using the exponential smoothing method? Assume the forecast for period 5 is 14000. Use a smoothing constant of α = 0.4 (Choose the closest answer.) a. 14575 b. 26100 c. 16600 d. 19700 C PTS: 1 11. The equation for a simple linear regression that saw sales averaging $225,000 over the last ten periods, and advertising budgets averaging $3,000 over the last 10 periods is: Y = 3250 + 120x This indicates that a $1 increase in advertising will increase sales by: a. $3370 b. $250 c. $120 d. $1875 C PTS: 1 12. Which one of the following is not a type of qualitative forecasting? a. Sales force composite b. Consumer survey c. Jury of executive opinion
d. Naïve method D PTS: 1 Data Set E2 Month Actual Forecast 1 10 11 2 8 10 3 9 8 4 6 6 5 7 8 13. A forecasting method has produced the following data over the past 5 months shown in Data Set E2. What is the mean absolute deviation (accurate to 2 decimals)? a. 0.60 b. 1.20 c. 1.00 d. 1.25 C PTS: 1 14. Based on the information in Data Set E2, what is the mean squared error (accurate to 2 decimals)? a. 7.00 b. 1.40 c. 1.00 d. 0.80 B PTS: 1 15. Using the actual demand shown in the table below, what is the forecast for May (accurate to 1 decimal) using a 4-month weighted moving average and the weights 0.1, 0.2, 0.3, 0.4 (with the heaviest weight applied to the most recent period)? a. 44.4 b. 43.0 c. 42.5 d. 41.6 Nov. Dec. Jan. Feb. Mar. Apr. 39 36 40 38 48 46 A PTS: 1 16. Given the following information, calculate the forecast (accurate to 2 decimals) for period three using exponential smoothing and α = 0.3. a. 36.90 b. 57.50 c. 61.50 Period Demand Forecast 1 64 59 2 70
d. 63.35 D PTS: 1 SHORT ANSWER 1. List and describe two types of qualitative forecasting methods. a. Jury of Executive Opinion: This type of qualitative forecasting is usually used to for longrange forecasting. An experienced group of senior management executives knowledgeable about the market, competitors, and the business environment develop a forecast in the hopes that their experience will provide them with a competitive advantage. b. Delphi Method: Internal and external experts are surveyed about future events and longterm forecasts of demand. Following the survey, the answers are summarized, and this summary is sent back out to the experts. Experts are then able to modify their responses based on the summary. The process continues until a consensus is reached. Experts in this process never physically need to meet; they communicate only through their survey responses and their subsequent responses to the summaries. c. Sales Force Composite: This qualitative forecasting method utilizes will 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. Based on this knowledge the sales force is asked to create a forecast to estimate the needs of the customer. d. Consumer Survey: Surveys are administered to customers via phone, mail, Internet, or personal interviews. Customers are surveyed and issues such as future buying habits, new product ideas, and opinions about existing products. Using statistical tools and managerial judgment, a forecast is devised. PTS: 1 2. The four components of time series data are: trend variations, cyclical variations, seasonal variations, and random variations. Briefly describe each type of variation. a. Trend variations: Trend variations can either be increasing or decreasing demand movements over a number of years. These variations can be due to factors such as population growth, population shifts, cultural change, and income shifts. b. Cyclical variations: This type of variation is typified by wave-like movements where demand fluctuates up-and-down. These fluctuations must be sustained for leased one year. Variation is usually influenced by macro economic and/or political factors. c. Seasonal variations: variation is seasonal; demand becomes somewhat predictable on certain days, weeks, months, years, or seasons. In the United States the most common form of seasonal variation is seen in many retail industries during the Christmas season when demand usually spikes. d. Random variations: Unexpected or unpredictable events such as natural disasters, strikes,
PTS: 1 and wars will cause random variations in demand. The threat of hurricane will usually cause a tremendous spike in demand for items like batteries, bottled water, and wood products. 3. List FOUR benefits that can be achieved by implementing a successful CPFR program. a. Strengthened partner/alliance relationships b. High-quality analysis of sales and order forecasts both upstream and downstream in the supply chain c. The ability to access and use accurate and timely data to improve forecast accuracy d. Demand change management with the ability to proactively eliminate problems before they appear e. Collaboration of future requirements and plans f. The integration of planning, forecasting, and logistics activities g. Improved effectiveness in category management and the ability to better understand consumer purchasing habits. h. Improved analysis of key performance metrics in the hopes of improving supply chain efficiency, improving customer service, increasing sales and profitability. PTS: 1 4. What is a tracking signal? How can managers use the information provided by the tracking signal to improve the quality of forecasts?? Tracking signal is used to determine if a forecasting model is stable over time. 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. The tacking signal provides an indication of whether the forecasts are consistently over or under actual demand. A biased forecast will lead to excessive inventories or stock-outs. Adjustments to the forecasts can be made accordingly. Some inventory experts suggest using ±4 for high-volume items and ±8 for lower-volume items while others prefer a lower limit. For example, a company may start off with a control limit for their tracking signal of ±4. Over time, the quality of forecasts improved and the control limits were reduced to ±3. As tighter limits are instituted there is a greater probability of finding exceptions that require no action, but it also means catching changes in demand earlier. PTS: 1 ESSAY 1. Use the data set below (Data Set E3) to answer the questions that follow.
Data Set E3 Period Sales Volume 1 1000 2 1200 3 1450 4 1750 5 2200 6 2750 a. Find the four-period simple moving average forecasts for Periods 5 and 6. b. Find the four-period weighted moving average forecasts for Periods 5 and 6 using weights of 0.05, 0.15, 0.30, and 0.50 from the earliest period to the latest period, respectively. c. Which set of forecasts is more accurate, the simple moving average forecasts or the weighted moving average forecasts? Why is that set of forecasts more accurate in this particular case (using Data Set E3)? d. Will that type of forecast always be more accurate? Why or why not? a. Find the four period simple moving average forecasts for Periods 5 and 6. Forecast for Period 5 = 1350 Forecast for Period 6 = 1650 b. Find the four period weighted moving average forecasts for Periods 5 and 6 using weights of 0.05, 0.15, 0.30, and 0.50 from the earliest period to the latest period, respectively. Forecast for Period 5 = 1540 Forecast for Period 6 = 1902.5 c. Which set of forecasts is more accurate, the simple moving average forecasts or the weighted moving average forecasts? In this case the weighted moving average forecasts are more accurate. One can use the mean absolute deviation (MAD) to compare forecast accuracy. For the SMA, the MAD for the 2 periods is 975. For the WMA forecast, the MAD is 753.8. Why is that set of forecasts more accurate in this particular case (using Data set E2)? Because demand is increasing in every period in Data Set E3 the weighted moving average forecast, which weighs more heavily recent periods of demand in calculating the forecast versus early periods of demand, is more capable of reflecting recent changes in demand patterns. d. Will that type of forecast always be more accurate? Why or why not? While the weighted moving average forecast works very well with this particular Data set, it might not be as effective if demand were more stable from period to period. When there's a greater degree of stability in demand from period to period, a simple moving average forecast would likely be more effective. Nonetheless, if demand patterns were very random it is uncertain which forecasting model would be more accurate. It is for this reason the textbook describes a number of different forecasting methods that when used together can help an organization see a more complete picture of what the future may hold. Use of the MAD is necessary to compare forecast accuracy.
PTS: 1 The student may also note that while the weighted moving average forecast was more accurate, neither forecasting model would have aptly prepared the organization for the increasing demands that were experienced during periods five and six. The student might even mention some of the other models from the textbook. The exponential smoothing forecasting model and trend adjusted exponential smoothing forecasting model may be more effective for this particular Data set. 2. You work for an auto parts manufacturer. Your manager has heard about CPFR (Collaborative Planning, Forecasting, and Replenishment) and requested you investigate the possibility of implementing CPFR in the near future. As part of your report you would need to address the following: a. Define CPFR? b. The Voluntary Interindustry Commerce Standards (VICS) Association has published a report outlining a model for CPFR implementation. What are the 8 collaborative tasks included in the VICS model? Is this a good model to follow? c. What are the performance metrics that can be used to measure success of the CPFR implementation? d. What are the key challenges of CPFR implementation? a. Define CPFR? According to the Voluntary Interindustry Commerce Standards (VICS) Association, "collaborative planning, forecasting and replenishment (CPFR) is a business practice that combines the intelligence of multiple trading partners in the planning and fulfillment of customer demand. CPFR links sales and marketing best practices, such as category management, to supply chain planning and execution processes to increase availability while reducing inventory, transportation and logistics costs." b. The Voluntary Interindustry Commerce Standards (VICS) Association has published a report outlining a model for CPFR implementation. What are the 8 collaborative tasks included in the VICS model? Is this a good model to follow? Task 1: Collaboration Arrangement (original Step 1) Task 2: Joint Business Plan (original Step 2) Task 3: Sales Forecasting (original Step 3) Task 4: Order Planning/Forecasting (original Step 6) Task 5: Order Generation (original Step 9a) Task 6: Order Fulfillment (original Step 9b) Task 7: Exception Management (original Steps 4, 5, 7, and 8) Task 8: Performance Assessment (new) VICS has been at the forefront of the CPFR movement. Many companies have successfully implemented CPFR using this model. An example is West Marine described in the text. c. What are the performance metrics that can be used to measure success of the CPFR implementation? The more common performance metrics are gross margin percent, return on investment, and sales growth. Other metrics include in-stock percent at point of sale, inventory turnover, inventory level, sales forecast accuracy, potential sales lost due to stock-out,
manufacturing cycle time, order cycle time, shipping cycle time, problem resolution time, rate of emergency or cancelled orders, and percent shipped or delivered on time. d. What are the key challenges of CPFR implementation? The key challenges are the difficulty of making internal changes, total implementation cost, and trust. As is true with any major implementation, internal resistance to change must be addressed by top management. Change is always difficult; however, if top management is committed to the project, then the project is much more likely to succeed. Companies will need to educate their employees on the benefits of the process changes and the disadvantages of maintaining the status quo. Although cost is an important factor, companies with no plans for adopting CPFR should determine whether they are at a competitive disadvantage, as more and more companies implement CPFR. Trust, a major cultural issue, is considered a big hurdle to widespread implementation of CPFR because many retailers are reluctant to share the type of proprietary information required by CPFR. Lack of trust is often associated with unreliable data and the lack of integration internal to suppliers and manufacturers. The real challenge to widespread adoption of CPFR is that it requires a fundamental change in the way buyers and sellers work together. PTS: 1