Question

g Opeai o ppy Chtheett 303 CASE Rachel's Breakfast Cafe Rachel Kirkptrick thought to berselif, "What a waste" an she threw away theee begs full ingredientsi have so gt beter at estimating how mach food to order and prepare. Rachel owned and operated a small cafe that special- ized in fresh baked quiches, beeakfast casseroles, andi breads, as well as ready-made country style beeakfasts The cale was open six days a wock and closed on Sundays. Rachel had run ber shop Sor over a year now, and business seemed to be taking off. While she had made a numiber of operating improvements related to the consistency and. quality of her reoducts she still struggled betwees two extremes of the same problem. On some days Rachel did not have enough ingredients on hand to satisty the day's customers. In this case, some of the people who espected to get one of her famous qaiches were disappointed. Other days Rachel had far too mach food on hand. On days liker today, Rachel found herself throwing away food because she had vastly overestimated the number of customers she would have. waold ims and speiled e her sopeles, he knew at he wold eed to weather formcast at leant this far in abvan e of me to her. Ai the same tim, he e serm forecasts are aally mre acoure new tha also decided to track the 2 day fec made two days in advance the that ieca Probahiity of Rain (3) 5 day foeseast Probabity of Rale( 3 day forat Tatail Day Monday Tuesday Wednesdays 0 40 20 30 4320 30 10 4312 Thurnday Priday 5545 Satarday 90 60 Monday 4590 Tuesday 30 4733 70 While it was difficult for Rachel to know exactly how many customers she served each day, she was able to accu- rately track the total dollar value of sales. It seemed to her that her business was growing, but she had not taken the time to see whether the increasing demand was a true trend or just her perception. Based on her assumption that busi- ness was growing, Rachel had been placing larger orders with ber suppliers of milk, eggs, cheese, and other ingredi- ents each week, Each day Rachel placed orders for supplies 90 Wednesday Tharsday 4923 50 4687 100 5988 100 100 Friday 20 6132 Saturday 10 5324 Monday 10 4526 10 Tuesday Wednesday 10 10 5232 10 50 5684. 50 online at a nearby grocer's website, and he delivered each order five days later. As Rachel considered how to improve her forecasts for needed items, she thought about possible factors that caused demand to be greater or smaller cach day at her café. Fridays and Saturdays were usually busier than other days. Beyond the weekend effect, she noted that the weather had an impact. On rainy days people were less likely to go out for breakfast. Rachel wondered how she could use this information to improve her business. Over the next four weeks Rachel collocted the data shown in the following table. The "5 day forecast" col- umn shows the probability of rain (percentage) for the area around Rachel's café, as predicted by the local weather service five days into the future. For example, the table shows a forecasted 10 percent probability of rain on the first Monday in the table; this was the forecast released by the weather service on the Wednesday five days earlier Since it currently took Rachel five days to receive orders Thursday 70 5911 20 Friday 6328 60 60 Saturday 4932 20 20 Mooday 5235 15 15 Tuesday 50 5862 20 Wedsesday 4862 20 20 Thursday 6100 1C Friday 6255 70 Saturday Questions 1. of the series of rain forecasts, respectively. Calculate the MPE (bias), MAD, and MAPE for the two fore- casting models. Which rain forecast seems to be beter Develop forecasts using regressions of sales on each 439 304 Supply Chain & Operation Management at predicting Rachel's daily sales, the five-day forecast or the two-day forecast? 2. How can Rachel make use of the rain forecasts to improve her forecasts of total sale other changes to her business would she need to make in order to capitalize on this information? 3. How are order lead time and forecasting accuracy related to each other in this case? Plot and visually inspect the sales data. What other suggestions would you give Rachel for improving her sales forecasts? What type of time series model would be appropriate? Why? 4. each day? What CASE C&F Apparel, Inc. Bill Smith, director of business planning for C&F Apparel, developed by Bill and his team were sometimes fairly accu- rate, forecasts for specific items were all over the map. Bill knew that their forecasting process was not as consistent he thought. Forecast errors for the fall season's sales from season to season as it could be, but he felt that flex- ibility was needed to cope with changing conditions. Bill had recently heard about "fast fashion" apparel that designs, pro- chewed on a fourth-story office. These bad forecasts are killing us, pencil as he looked out the window of his had ranged from 50 to 200 percent of demand. As a avily

Answer 1

forecat of ales s the Folleasing forecast: on Aeries of yain C D Tota 2 day forecast 5 day forccast Day Sales 40 55a0 10 Monday 2 4320 30 20 Tuesday 30 Wednesday 4987 50 5 thurday 80 5545 nday Acturday 60a3 90 4590 30 8 Mon T33 30 70 49a3 9Tue 30 o Wed 50 4687 loo u Thuy 5988 6/3& 2 fi 20 13 Sat 53a4 4586 Mon 1C s Tue 1C 0 5684 6 Wed 50 7Thu 50 5911 20 63&8 60 Sart 60 4932 20 20 Mon 20 5335 15 21 Tue 15 5868 50 22 Wed 20 486a 20 23 Thur 20 6100 80 6955 70 60

Day FoTecast Slepe SLOPE (Total (Atae) Totol (End 5 Day Fotecast (Start) 5 day foreas Takercet Total (End INTERCEPT (Total (stert 5 day Forecast (Slart): 5 day forecact(d ( Stepe D2: D 25 B2: B25 SLOPE - keepts 5,255.45 INTERCEPT B2: B25 D2: D2S Day Forecast(In same oay 2 SLOPED2: D 25 7.73 Stope C2 C25 cas) D2: D25 C2:C25 Jakercapt INTERCEPT 4,542. 85

Book Microsoft Excel Home Insert Page Layout Formulas Data Review View Cut Calibri Wrap Text E AutoSu FE Copy General oacte Format Painter Clipboard BIU Merge &Center Conditional Format Formatting as Table tosert Delete Formal Sort&Find Clear Fi Seld Cen Stytes Font Allignment Number Styies Cells M14 D Regressian voalues Total Regression Modutus vafues A-F/F 5 Day forecast Total sales Day Forecast 5 Day Forer 2 Day Forecast 5 Doy forecast 2 Day forecact 5 Day forecost 2 Day Forecast 5 Day 2 Day 40 5520 5272 8 0.046872185 0051018 247 15 267.95 4 970 25 30 4320 5290 25 5074.75 306428457-0107655 10 712 5307 65 4720 15 s08.15 1095.65 1095 65 508.15 6- 50 an 5342 45 0066533145 0.050466 0.027870205 0069826 0112886984 0018824 a.143632751 0095522 0.119810312 0.067343 5252.05 265.05 355.45 855 45 265.05 80 80 5545 15035 610.95 5394 55 150 35 5961 25 416.25 416.25 5023 5412 05 9u 90 6138.55 610.95 115.55 115.55 4590 484.75 5359 85 5074.75 769.85 769.85 484.75 10 341.75 644 25 341 75 11 90 100 4923 30 5412.05 29903 a z47 45 4687 5429.35 5429.45 742.35 742 45 742-35 13 100 5988 I00 5429.45 6315.85 327.85 .10287414 0.051909 558.55 558.55 327.85 14 5290 25 841 751 20 6132 348.05 603.85 70 5783.95 348 05 84175 0.159113463 0060175 5324 4720 15 15 TO 5272 85 51.15 603.85 o 009700636 5115 0.12793 746 853 5272 85 16 4526 0.141640669 0.041132 a.007747233 0.1084397 0.063931342 0.046902 0.1173385 0021966 10 4720:15 194.15 746 85 194.15 40.851 17 12. 5272 85 5232 4720.15 51185 10 10 40.85 511.85 5429 35 50 341.55 5/ 5684 5342.45 754.55 341 55 254.65 19 5290.25 5783.95 620.75 127.05 620.75 50% 5911 127.05 6328 5359.85 60 0.180630055 0.12866 0.067718917 0.007055 5606.65 968.15 721.35 20 958.15 721 35 60 4897 45 4932 5290 25 358.25 34.55 358.25 1 34.55 20 20 15 15 8813701 0.088629 571 75 432.65 571 75 20 50 4862 6100 6255 0080950806 0007238 0.156869625 0.023275 0.167010271 0.081441 489745 -428.25 35.45 428.25 35 as 5290.25 s272 85 5359.85 827.15 138,75 827 15 138.75 5961 25 471 05 895 15 895.15 S78a.95 471.05 361 4895833 557.1708333 0104456715 0.068769 Steeti Sheet2Sheeta

Regresrion intereyt b Slope R = a+ bx . 4 (10) 5255. 45 Sales X 5272. 82 Values Ragesson Columns in Dand (E Total Sales A F PBgesc ression aSe in Cdlemns P and G

Columm in Total CAF) Column (A-F 1 MFE= EA-F A-F Values are in olumns and 5-Day forecast, MAD |A-Fl FoY n in Column(H Average (H28) 557. 17 for 2-Day foccast MAD 드 EA-FI in Column (I Avernge 361. 4895 I 28 AFlues and (k in Columns are A-F For MAPE 5-Day Forecasé/ x 100 n inColumn (J) xl00 Avernge O. 1044 X 00 Crox I0. 4 For 2 o.0687 X loo MA PE Day Forecast 6. &Y )

2 Day ForEcast Day Forecast O.00 MFE 557.7 361.4895 MAD MAPE O. 10 O. OT