# Forecasting Problem 1 (10 points)

 Forecasting Problem 1 (10 points) The following equation summarizes the trend portion of quarterly sales of automatic dishwashers over a long cycle. Sales also exhibit seasonal variations. Ft = 40 – 6.5t + 2t2 where Ft = Unit sales (in 000 units) t = 1 at the first quarter of 2010 Quarter Relative 1 110 2 100 3 60 4 130 a)   Using the information given, prepare a seasonalized forecast of sales for each quarter of 2014. Year Qtr t Seasonally Unadjusted Forecast Quarterly Index Seasonally Adjusted Forecast 2014 1 17 507.50 110% 558.25 2 18 571.00 100% 571.00 3 19 638.50 60% 383.10 4 20 710.00 130% 923.00 b)   Prepare a quarter-by-quarter time series plot showing the forecast trend (seasonally unadjusted forecast) from the first quarter of 2010 to the fourth quarter of 2014. Superimpose on this graph the seasonally adjusted forecast for the same time period. Year Qtr t Seasonally Unadjusted Forecast Quarterly Index Seasonally Adjusted Forecast 2010 1 1 35.50 110% 39.05 2 2 35.00 100% 35.00 3 3 38.50 60% 23.10 4 4 46.00 130% 59.80 2011 1 5 57.50 110% 63.25 2 6 73.00 100% 73.00 3 7 92.50 60% 55.50 4 8 116.00 130% 150.80 2012 1 9 143.50 110% 157.85 2 10 175.00 100% 175.00 3 11 210.50 60% 126.30 4 12 250.00 130% 325.00 2013 1 13 293.50 110% 322.85 2 14 341.00 100% 341.00 3 15 392.50 60% 235.50 4 16 448.00 130% 582.40 2014 1 17 507.50 110% 558.25 2 18 571.00 100% 571.00 3 19 638.50 60% 383.10 4 20 710.00 130% 923.00

Number of Dishwashers Sold

2010-2014

Seasonally Unadjusted Forecast 2010.0 2011.0 2012.0 2013.0 2014.0 35.5 35.0 38.5 46.0 57.5 73.0 92.5 116.0 143.5 175.0 210.5 250.0 293.5 341.0 392.5 448.0 507.5 571.0 638.5 710.0 Seasonally Adjusted Forecast 2010.0 2011.0 2012.0 2013.0 2014.0 39.05 35.0 23.1 59.8 63.25000000000001 73.0 55.5 150.8 157.85 175.0 126.3 325.0 322.85 341.0 235.5 582.4 558.25 571.0 383.1 923.0

## Fore Prob 2

 Forecasting Problem 2 (10 points) The data below represent the relative shares (by quarter) of call volumes over 16 quarters from a call center at a major financial institution. 2010 2011 2012 2013 Average Qtr Percent Share Q1 23.2% 23.0% 23.3% 21.9% 22.8% Q2 25.1% 24.6% 26.2% 25.3% 25.3% Q3 28.5% 28.8% 28.6% 29.8% 28.9% Q4 23.2% 23.6% 21.9% 23.1% 22.9% Total 100% 100% 100% 100% 100% (a) Using the average quarter percent share column (Column G), generate a pie chart and a bar chart to show the quarterly percent shares of call volumes to the call center. (b) Using the percentage table above, what are the indices for each of the four quarters? Index Q1 91.3% Q2 101.2% Q3 115.7% Q4 91.8% 400.0% (c) Assume that the projected number of calls for the year 2014 is 50,000,000, what are the seasonally adjusted forecasts for the number of calls for Q1, Q2, Q3, and Q4? Forecast Number of Calls 2014 50,000,000 Forecast, By Quarter Q1 11,413,027 Q2 12,651,833 Q3 14,465,680 Q4 11,469,460 50,000,000

Typical Volume of Call Center, Percent Distribution by Quarter

Average Qtr Percent Share

0.228260546938608 0.253036651885006 0.289313598547655 0.229389202628731

Typical Volume of Call Center, Percent Distribution by Quarter

Average Qtr Percent Share

0.228260546938608 0.253036651885006 0.289313598547655 0.229389202628731

Typical Volume of Call Center, Percent Distribution by Quarter

Average Qtr Percent Share

0.228260546938608 0.253036651885006 0.289313598547655 0.229389202628731

Typical Volume of Call Center, Percent Distribution by Quarter

Average Qtr Percent Share

0.228260546938608 0.253036651885006 0.289313598547655 0.229389202628731

## Fore Prob 3

 Forecasting Problem 3 (15 points) The data below represent the call volumes over 16 quarters from a call center at a major financial institution. Develop a forecasting model for the volume of calls (in 000 units). 2010 2011 2012 2013 Q1 473 544 628 709 Q2 513 582 707 725 Q3 582 681 773 854 Q4 474 557 592 661 (a) Create a time series graph showing the: (1) actual data, (2) trend line for the data, and (3) deseasonalized actual data. Label the graph appropriately. Year Qtr t Actual Volume of Calls Deseasonalized Actual Volume of Calls 2010 1 1 473 483.18 2 2 513 503.68 3 3 582 515.76 4 4 474 542.25 2011 1 5 544 555.71 2 6 582 571.43 3 7 681 603.49 4 8 557 637.20 2012 1 9 628 641.52 2 10 707 694.16 3 11 773 685.02 4 12 592 677.23 2013 1 13 709 724.27 2 14 725 711.83 3 15 854 756.80 4 16 661 756.17 (b) Develop the quarterly (Q1, Q2, Q3, Q4) indexes for the volume of calls See page117-118 of text for similar problem Year Qtr t Volume of Calls MA4 MA2 Y/MA2 2010 1 1 473 2 2 513 510.5 3 3 582 528.25 519.375 1.120578 Unadjusted Index Adjusted Index 4 4 474 545.5 536.875 0.882887 Qtr 2011 1 5 544 570.25 557.875 0.975129 1 97.8% 97.9% 2 6 582 591 580.625 1.002368 2 101.7% 101.9% 3 7 681 612 601.5 1.132170 3 112.7% 112.8% 4 8 557 643.25 627.625 0.887473 4 87.3% 87.4% 2012 1 9 628 666.25 654.75 0.959145 399.5% 400.0% 2 10 707 675 670.625 1.054240 3 11 773 695.25 685.125 1.128261 4 12 592 699.75 697.5 0.848746 2013 1 13 709 720 709.875 0.998767 2 14 725 737.25 728.625 0.995025 3 15 854 4 16 661 a 473.68 (c ) Using the trend line you developed in Part (a), what are your seasonally unadjusted and seasonally adjusted forecasts for the four quarters of 2014? b 18.207 Year Qtr t Seasonally Unadjusted Forecast Quarterly Index Seasonally Adjusted Forecast 2014 1 17 783.199 97.9% 766.69 2 18 801.406 101.9% 816.23 3 19 819.613 112.8% 924.88 4 20 837.820 87.4% 732.37

Volume of Calls to Call Center 2010 to 2013

Actual Volume of Calls

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2010 2011 2012 2013 473.0 513.0 582.0 474.0 544.0 582.0 681.0 557.0 628.0 707.0 773.0 592.0 709.0 725.0 854.0 661.0 Deseasonalized Actual Volume of Calls 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2010 2011 2012 2013 483.1849331666253 503.6807630447924 515.7593318992456 542.2452597374343 555.713749773032 571.4272984250861 603.4915893872617 637.195379058546 641.522490546809 694.1565291864878 685.0205559417816 677.2345860011836 724.2666334358083 711.8295384161296 756.8014938865219 756.1690225452405 0.0 Linear 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 2010 2011 2012 2013 1.0

## Fore Prob 4

 Forecasting Problem 4 (15 points) Many supply managers use a monthly reported survey result known as the purchasing managers’ index (PMI) as a leading indicator to forecast future sales for their businesses. Suppose that the PMI and your business sales data for the last 10 months are the following: 1 2 3 4 5 6 7 8 9 10 Month PM 43 43.1 41.5 38.5 40.5 45.2 46.2 48.1 49 53 Sales (in \$000) 122 124 125 123 119 120 125 127 135 136 A.     Develop a regression model that can be used by supply managers in forecasting future sales for businesses. Explain what forecasting model approach you used and why you chose it. Show complete work (cut and paste from Excel if used in the analysis). (10 points) B.     Develop a sales forecast for the 11th and 12th months using the model you developed in part A when PMIs are 52 and 50, respectively. (10 points) Y= 1.0529x + 78.421 = 50 = 131.066 52 = 133.1718

PMI Vs. Sales

Sales (in \$000)

43.0 43.1 41.5 38.5 40.5 45.2 46.2 48.1 49.0 53.0 122.0 124.0 125.0 123.0 119.0 120.0 125.0 127.0 135.0 136.0

PMI

Sales (in 000)

## Fore Prob5

 Forecasting Problem 5 (15 pts) An electrical contractor’s records during the last 5 weeks indicate the number of job requests: Week Actual Requests 1 20 2 22 3 18 4 21 5 22 a) Graph the actual request data using appropriate labels, and provide insights about the time series (describe what you observe re the behavior of the sales over the period under review). b) What is the forecast for Week 6 using a 2-period moving average? 21.5 c) What is the forecast for Week 6 using the Naïve method? 22 Compute the MAD, MAPE, and MSE for the two-period moving average and Naïve models and compare your results. Explain which of the two forecasting models you prefer and why. d) Forecasts MAD MSE MAPE Week Actual Requests MA2 Naïve MA2 Naïve MA2 Naïve MA2 Naïve 1 20 2 22 20 2 4.000 9.09% 3 18 21 22 3 4 9.000 16.000 16.67% 22.22% 4 21 20 18 1 3 1.000 9.000 4.76% 14.29% 5 22 19.5 21 2.5 1 6.250 1.000 11.36% 4.55% MAD 2.2 2.5 MSE 8.1 10.0 MAPE 10.93% 12.54% e) Graph the actual number of requests, the 2-period and Naïve forecasts. Use appropriate labels for your graphs

Actual Requests

Actual 20.0 22.0 18.0 21.0 22.0

Actual Requests Actual vs. MA2 and Naive Forecasrs

Actual Requests 20.0 22.0 18.0 21.0 22.0 MA2 21.0 20.0 19.5 Naïve 20.0 22.0 18.0 21.0

Actual Requests

Actual 20.0 22.0 18.0 21.0 22.0

Actual Requests Actual vs. MA2 and Naive Forecasrs

Actual Requests 20.0 22.0 18.0 21.0 22.0 MA2 21.0 20.0 19.5 Naïve 20.0 22.0 18.0 21.0

## Rel Prob 1

 Reliability Problem 1 (15 points) One of the industrial robots designed by a leading producer of servomechanisms has three major components. Components’ reliabilities are 80, 85, and 95%. All of the components must function in order for the robot to operate effectively. a.  Compute the reliability of the robot. 0.8 0.85 0.95 Reliability = 0.646 b.  Designers want to improve the reliability by adding a backup component. Due to space limitations, only one backup can be added. The backup for any component will have the same reliability as the unit for which it is the backup. Which component should get the backup in order to achieve the highest reliability? Show proof of your answer by computing the overall reliabilities of the three options (assume 100% reliable backup switch) 0.96 Tring all components with the formula above, we got component with 80% reliability c.   If one backup with a reliability of 99% can be added to any of the main components, which component should get it to obtain the highest overall reliability? Show proof of your choice by computing the overall reliabilities of the three options (assume a backup switch with 100% reliability). Switch with 1 0.8 0.99 0.99800 0.998 0.85 0.99 0.99850 0.9985 0.95 0.99 0.99950 0.9995 So, Component with 80% reliability

## Rel Prob 2

 Reliability Problem 2 (10 points) Lucky Lumen light bulbs have an expected life that is exponentially distributed with a mean of 20,000 hours. Determine the probability that one of these light bulbs will last: a.       At least 24,000 hours t 24000 MTBF 20000 P(T>=24000) 0.3011942119 b.       No longer than 4,000 hous t 4000 MTBF 20000 P(T<4000) 0.8187307531 Faliure Rate 1-P(T<4000) 0.1812692469 c.       Between 4,000 and 24,000 hours t 24000 t 4000 MTBF 20000 MTBF 20000 P(T>=24000) 0.3011942119 P(T<4000) 0.8187307531 0.517537

## Rel Prob 3

 Reliability Problem 3 (10 points) An office manager has received a report from a consultant that includes a section on equipment replacement. The report indicates that scanners have a service life that is normally distributed with a mean of 41 months and a standard deviation of 4 months. On the basis of this information, determine the percentage of scanners that can be expected to fail in the following time periods. a. Before 38 months of service. t 38 Mean 41 SD 4 Z -0.75 P(T<38) 22.66274% b. Between 40 and 45 months of service. t 40 45 Mean 41 41 SD 4 4 Z -0.25 1 P(40

## Rel Prob 4

 Reliability Problem 4 (10 points) How high must reliability be? Prime business customers expect public carrier-class communications data links to be available 99.999 percent of the time. The so-called five nines rule implies only 5 minutes of downtime per year. Such high reliability is needed not only in telecommunications but also for mission-critical systems such as airline reservation systems or banking fund transfers. Suppose a certain network web server is up only 90 percent of the time (i.e. its probability of being down is 0.10). How many independent servers are needed to ensure that the system is up at least 99.999 percent of the time? Show your work and explain your answer. Server Rel. 90% Machine Failure Rate 10% Number of Servers 1 2 3 4 5 6 P(failure) 0.100000 0.010000 0.001000 0.000100 0.000010 0.000001 P(Reliable) 0.900000 0.990000 0.999000 0.999900 0.999990 0.999999 % 90 99 99.9 99.99 99.999 99.9999