need in 1 hour
budsimpson
Week6DQ2.docxWeek 6 DQ's/Lecture- Chap 7
Jesse
DQ# 1: There are two types of demand. What are they, and how do they influence the supply chain?
There are two types of demand: independent and dependent. Independent simply means demand for an end-item. For example, demand for cars (let's say Teslas) is an independent demand. Dependent means that the demand for a given item is dependent upon demand for another item (Coyle, et al). For example, demand for recharging centers for an electric cars (such as Teslas) is dependent upon the demand for cars.
This has implication for the supply chain, as the forecast for dependent items is contingent upon the forecasted demand for independent items. Therefore, we can expect an inverse relationship between dependent and independent demand items. Repair parts are a good example of how the production of dependent-demanded items are directly proportional to independent demand. As demand for a specific independent item (like a Tesla) goes up, so does the demand for all items that is required to support the consumption of a Tesla, like recharging stations, batteries or repair parts. There will be no demand for recharging stations without a demand for Teslas. This impacts the demand forecast, as models can better predict the demand for a dependent item based on the demand for the independent item.
DQ# 2: There are at least three forecasting methods. Name them and choose one to discuss in more detail, including advantages and disadvantages.
Three forecasting methods are a simple moving average, weighted moving average and exponential smoothing.
A simple moving average is simply the average demand over a preset number of periods and uses this average as the demand for the next period. It's use is very simple, as it is a simple average over a given period of time. It can also give you a forecast very quickly, lending to its ease of computation. However, it assumes that all the historical demand are weighted the same, so the analyst cannot assign any further granularities to the inputs of the forecast beyond a simple average.
Weighted moving average takes this concept one step further by giving different "weights' to different periods. Weighted moving average gives a greater weight to periods or variables that the forecaster believes is more important. Typically, data from recent  periods are given greater weight, but weights can be assigned based on any reason the forecaster believes is more representative of the future he/she is trying to forecast. A weighted moving average is easy to use and allows for greater manipulation of the data set. It also allows for greater tracking signal and less bias than a simple moving average.
Its disadvantage is the fact that it assumes that behaviors in the past will hold in the future. It is a projection based on the future based on past observations (Murphy and Knemeyer, 2015). As such, a weighted moving average is based on assumptions about the past that is extrapolated in some manner to draw conclusions about the future. Each of these assumptions carries risk. This risk is that the assumption that was true in the past will also hold true for the future. This may or may not be the case. By assigning a weight to certain values, we control this extrapolation of the past so that only variables that we believe to be the most prudent will be considered by the forecast. Weighted averages attempts to hedge against future risks based upon probability (such as assigning past values a weight) but no measure will be able to account for all randomness inherent within each forecast. However, the weight we assign to a value is only as good as one's judgment, which may or may not hold to be accurate regardless of how experienced we may be. This is why there will always be a delta between the forecast and reality.
Coyle, J.J., Langley, C.J., Novack, R.A., & Gibson, B.J. (2017). Supply Chain Management: A Logistics Perspective. (10th ed.). Boston, MA: Cengage Learning.
Murphy, Jr, P.R. and Knemeyer, A.M. (2015). Contemporary Logistics. Saddle River, NJ: Pearson Education, Inc
