# Operation management

Assignment 1

1. Sales data for two years are as follows. Data are aggregated with two months of sales (in 1,000 units) in each “period.”
 Year 1 Year 2 Period Sales Period Sales January–February 126 January–February 172 March–April 152 March–April 151 May–June 165 May–June 204 July–August 184 July–August 238 September–October 167 September–October 189 November–December 123 November–December 149
1. Plot the data.
2. Fit a linear regression model to all the sales data.
3. In addition to the regression model, determine multiplicative seasonal index factors. A full cycle is assumed to be a full year.
4. Using the results from parts b) and c), prepare a forecast for the next year.
• Zeus Computer Chips Inc. used to have major contracts to produce the Centrino-type chips. Here is demand over the past 12 quarters:
 Year 2016 2017 2018 I 5700 I 4400 I 3400 II 4000 II 3100 II 2800 III 4900 III 4400 III 2700 IV 3700 IV 3100 IV 2000

Fit all the data above by a linear regression model with an additive form (using dummy variables) to forecast the four quarters of 2019.

• The demand manager of Maverick Jeans is responsible for ensuring sufﬁcient warehouse space for the ﬁnished jeans that come from the production plants. In order to estimate the space requirements the demand manager is evaluating moving-average forecasts. The demand (in 1,000 case units) for the last ﬁscal year is shown below.
 Month 1 2 3 4 5 6 7 8 9 10 11 12 Demand 23 26 24 28 24 30 24 20 31 22 27 31
1. Use a three-month moving average to estimate the month-in-advance forecast of demand for months 4–12 and generate a forecast for the ﬁrst month of next year. Calculate mean absolute deviation (MAD).
2. Use an exponential smoothing method with a starting forecast of 21 for month 1 and a smoothing constant α = 0.5 to calculate month-in-advance forecasts for months 4–12 and forecast for the ﬁrst month of next year. Calculate the MAD.
3. Compare the MAD for the forecasting methods in parts a) and b). Based on these error calculations, which of the two forecast methods would you recommend?