Download Quiz 2 in ISyE 3103: Introduction to Supply Chain Modeling - Transportation and Logistics and more Quizzes Systems Engineering in PDF only on Docsity! Name: . . . . . . . . . . . . . . . . . . . . . . . . ISyE 3103 Introduction to Supply Chain Modeling: Transportation and Logistics Spring 2006 Quiz 2 Date: February 15, 2006 Instructions 1. There are 4 pages plus 1 formula sheet. 2. This part of the quiz is worth 40 points. Your grade here will be combined with that obtained from Homework 4 (maximum of 10 points) to comprise your total quiz grade. 3. Do your own work. 4. Show all calculations to ensure partial credit. Correct answers without any work will not receive full credit. 5. Give it your best shot! Question 1 (13 points) 1. Why were Barilla’s distributors opposed to the idea of vendor-managed inventory (VMI)? [5 points] Answer: The distributors thought that Barilla’s representatives were implying that they wanted to manage the inventories because the distributors were incapable of do- ing it themselves. They believed that they could manage their inventory levels and increase their service levels if Barilla could deliver their orders on time. They were also concerned that Barilla was looking for an excuse to push extra inventory into the distributors’ warehouses, and they were skeptical that Barilla was capable of managing the distributors’ inventories effectively. Some distributors were only willing to share their demand data if Barilla paid them for it. 2. List two of the potential dangers associated with category management initiatives. [4 points] Answer: Category captains’ recommendations could become pervasive so that every store stocks the same products in the same quantities, which limits consumers’ choices. Small vendors often find it difficult to gain notoriety with the captain; consequently, they may have difficulty obtaining shelf-space allocation. Captains have the incen- tive to stock more of their products and remove competitors’ products from the mix; allocation recommendations should be verified by another vendor. ISyE 3103 · Spring 2006 · Quiz 2 2 3. Discuss the importance of cost and accuracy in the process of evaluating various fore- casting methods. [4 points] Answer: More accurate forecasting methods are typically more costly to produce. Decision makers should identify the benefits of obtaining a more accurate forecast and determine the costs of forecast inaccuracy. Simple forecasting methods often produce very accurate results. More importantly, complex forecasting methods often cannot improve forecast accuracy significantly over the simpler methods. Question 2 (15 points) The following historical demand data for Blue Öyster Cult’s Cult Classics CD at Northeast Ohio FYE Stores is available: Month Jan 05 Feb 05 Mar 05 Apr 05 May 05 Jun 05 Demand 119 72 113 82 82 131 1. Fit a 3-period moving average model to the data above. Produce forecasts for as many historical periods as you can. Use your model to forecast the next three months’ (July, August, and September 2005) demand for the Cult Classics CD. [6 points] Answer: Month t Dt ft;t−1 et |et| January 1 119 February 2 72 March 3 113 April 4 82 101.33 −19.33 19.33 May 5 82 89.00 −7.00 7.00 June 6 131 92.33 38.67 38.67 The forecast for each of the next 3 months is f7;6 = f8;6 = f9;6 = D4 + D5 + D6 3 = 98.33. Recall that the multiple-step-ahead forecasts for the moving average model are equal to the one-step-ahead forecasts. ISyE 3103 · Spring 2006 · Quiz 2 5 f26;24 = ( â24 + b̂24(2) ) Î22 = 319.1304 f27;24 = ( â24 + b̂24(3) ) Î23 = 318.5867 f28;24 = ( â24 + b̂24(4) ) Î24 = 235.4908 Bonus (1 point each) Identify the meaning of the following supply chain acronyms: 1. EDI Answer: Electronic Data Interchange 2. ERP Answer: Enterprise Resource Planning 3. XML Answer: Extensible Markup Language ISyE 3103 · Spring 2006 · Quiz 2 6 Time Series Extrapolation Forecasting Formulas p-Period Moving Average Model ft+1;t = ∑t i=t−p+1 Di p = ft+τ ;t Exponential Smoothing Model with Smoothing Parameter α ft+1;t = αDt + (1− α)ft;t−1 = ft+τ ;t Holt’s Method (Double Exponential Smoothing) with Parameters α and β ât = αDt + (1− α) ( ât−1 + b̂t−1 ) b̂t = β (ât − ât−1) + (1− β)b̂t−1 ft+1;t = ât + b̂t ft+τ ;t = ât + b̂tτ Winters’ Method (Triple Exponential Smoothing) with Parameters α, β, and γ ât = α Dt Ît−m + (1− α) ( ât−1 + b̂t−1 ) b̂t = β (ât − ât−1) + (1− β)b̂t−1 Ît = γ Dt ât + (1− γ)Ît−m ft+1;t = ( ât + b̂t ) Ît+1−m ft+τ ;t = ( ât + b̂tτ ) Ît+τ−m Mean Square Error (MSE) MSE = ∑n i=1 e 2 i n Root Mean Square Error (RMSE) RMSE = √ MSE Mean Absolute Deviation (MAD) MAD = ∑n i=1 |ei| n Mean Absolute Percent Error (MAPE) MAPE = ∑n i=1 |ei| Di n