OPMA Chapter 3 Notes
OPMA Chapter 3 Notes OPMA 3306
Popular in Operations Management
Popular in Business, management
verified elite notetaker
This 4 page Class Notes was uploaded by Sarah Quinn on Thursday February 4, 2016. The Class Notes belongs to OPMA 3306 at University of Texas at Arlington taught by Dr. Michel E. Whittenberg in Spring 2016. Since its upload, it has received 115 views. For similar materials see Operations Management in Business, management at University of Texas at Arlington.
Reviews for OPMA Chapter 3 Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/04/16
OM Explorer is excel addin on blackboard. Slightly different formulas. Demand estimates for products and services are all starting points for all other forecasts in OM. Based in part on demand estimates. Inputs to both business strategy and production resource forecasts. Must use to plan. Other things can be forecasted too. Vital function and impacts every management decision. Different things need different forecasts. 4 basic types. Qualitative (educated guess), timeseries analysis, causal relationships, simulation. Time series is based on relating to past demand can be used to predict future. Uses of forecasting: if demand exists, long term capacity needs, midterm fluctuations in demand, shortterm fluctuation in demand. Components are average demand for period of time, trend, seasonal elements, cyclical elements (long periods of time, years), random variation, and autocorrelation. Trends include linear, s curve, asymptotic, and exponential. Quantitative based on past demand forces and patterns. Qualitative is judgmental. Useful for new products. Forecast errors are not the sequence presented in the textbook. Difference between what was forecasted and what actually occurred. Always gonna be wrong. Bias and random sources of error. 4 measures of error. MAD, MAPE, TS, Standard error of estimates (regression analysis, all others are timeseries). For MAD, want number as close to 0. Know Running sum of forecast error RSFE or CFE (Sigma Et). MAD is Sigma Et all over n. MAPE scales forecast error to magnitude of demand. MAPE = MAD/Average demand. TS=RSFE/MAD. Know MAD. Forecast is actual demand forecast demand. MAD is sum of absolute deviation divided by months. TS is RSFE divided by MAD. Actual demand is At or Dt. Average forecast error is CFE/number of months or periods. RSFE of 15 is slight bias to overestimate demand. Shortterm is tactical decisions, mediumterm to develop strategy, longterm to detect general trends. Choose forecast model based on time horizon, data availability, accuracy required, size of forecasting budget, and availability of qualified personnel. Simple moving average: each period is equally weighted, oldest data not used. More Average period, forecast less responsive to demand and vice versa. Removes some random fluctuation in data. Weighted moving average, time periods not equal. Weights must add to 1.0. Exponential smoothing requires actual demand, forecast of demand, and smoothing constant. Understand concept of exponential smoothing. Trend adjustment delta makes more accurate forecast. Simple linear regression used to identify relationship between 2 or more correlated variables (one dependent and one independent at least). Assures relationship can be explained with straight line. Y=a+bt. Coefficient of correlation is r. Measures direction and strength of relationship between indep and dep variables. From 1 to 1. Coeff of determination is r2. 0 to 1. Standard error of the estimate, Syt, or Syx. Lin reg is least squares analysis. Makes errors = 0. Will not do it manually. Causal can use other than time variables. r tells us direction and strength of the relationship. R2 is amount of change in dependent variable. Time Series Decomposition. Chronologically ordered data. Seasonal variation is main focus in class. Is ratio of amount sold during each season divided by average for all seasons. 1. Decompose time series into components, 2. find future value. 3. develop least squares regression line for deseasonalized data. 4. Project regression line through period of forecast. 5. Adjust regression line by seasonal factor. Multiple Regression Model. Additional independent variables NOT used in class. Qualitative Methods. Takes advantage of expert knowledge. Used when no data is had to support a forecast. Use market research, panel consensus, historical analogy (similar product used for new product forecast), delphi method (another type of panel consensus but with secret ballots). CPFR (look at this in textbook). Forecasts always wrong! Criteria for selecting timeseries method: minimize biases like RSFE or CFE, minimize MAD. Smoothing constant, initial forecast, actual demand for exponential smoothing method. Tracking signal = rsfe/mad. Forecasting Does play role as basis for supply chain planning. 4 basic types of forecasts. 4 primary uses of forecasting. Methods discussed for time series analysis (simple moving average, weighted moving average, exponential smoothing, exponential smoothing with trends, causal relationships.). 4 types of forecast errors (mad, mape, standard error of linear regression). 4 qualitative forecasting methods (market research, panel consensus, historical analogy, delphi method). Chapter 3: Forecasting Strategic forecasts: medium and longterm forecasts that are used for decisions related to strategy and aggregate demand. Tactical forecasts: shortterm forecasts used for making daytoday decisions related to meeting demand. Needed at decoupling points to set appropriate inventory levels for buffers. Perfect forecast virtually impossible. Types of Forecasting Qualitative, time series analysis, casual relationships, and simulation. Qualitative are subjective and based on estimates and opinions. Time series analysis: a forecast in which past demand data is used to predict future demand. Casual assumes demand is related to some underlying factor or factors in the environment. Simulation allows forecaster to run through assumptions about the condition of the forecast. Components of Demand Demand can be broken down into 6 components: average demand for the period, a trend, seasonal element, cyclical elements, random variation, and autocorrelation. Time Series Analysis Tries to predict future based on past data. Short is usually under 3 months, medium is 3 months to 2 years, and long is past 2 years. When forecasting, firm should choose model depending on time horizon to forecast, data availability, accuracy required, size of forecasting budget, and availability of qualified personnel. Simple Moving Average Moving average: a forecast based on average past demand. When demand is neither growing nor declining rapidly and no seasonal characteristics. Weighted Moving Average weighted moving average: a forecast made with past data where more recent data is given more significance than older data. Choosing Weights Trial and error and experience. Most recent past is most important indicator. Exponential Smoothing Exponential smoothing: a time series forecasting technique using weights that decrease exponentially for each past period. Smoothing constant alpha: the parameter in the exponential smoothing equation that controls the speed of reaction to differences between forecasts and actual demand. Trend Effects in Exponential Smoothing Smoothing constant delta: an additional parameter used in an exponential smoothing equation that includes an adjustment for trend. Choosing the Appropriate Value for Alpha and Delta Smoothing constants are given a value between 0 and 1. Typically small values. Linear Regression Analysis Regression is a functional relationship between two or more correlated variables. Used to predict one variable given another. Linear regression is special class of regression where relationship between variables forms a straight line. Y=a+bt. Y is value of dependent variable, a is Y intercept, b is slope, and t is index for time period. Useful for longterm forecasting or major occurrences and aggregate planning. Used for time series and casual relationship forecasting. Decomposition of a Time Series Time series is chronologically ordered data that may contain one or more components of demand. Decomposition: the process of identifying and separating time series data into fundamental components such as trend and seasonality. When demand contains seasonal and trend effects at the same time, how do they relate to each other? Additive Seasonal Variation Assumes seasonal amount is constant no matter what trend or average amount is. Forecast = trend + seasonal. Multiplicative Seasonal Variation Trend is multiplied by seasonal factors. Forecast = trend X seasonal factor. Seasonal Factor (or Index) amount of correction needed in a time series to adjust for the season of the year. Seasonal is a period of the year characterized by some particular activity. Cyclical for anything else. Decomposition Using Least Squares Regression 1. Decompose the time series into its components. a. Find seasonal component. b. Deseasonalize the demand. c. Find trend component. 2. Forecast future values of each component. a. Project trend component into the future. b. Multiply trend component by seasonal component. Error Range Usual errors similar to standard deviation of any set of data. And errors because the line is wrong. Forecast Errors: Forecast error: the difference between actual demand and what was forecast. Errors called residuals. Distinguish between sources of error and measurement of error. Sources of Error Common source is projecting past trends into the future. Bias errors and random errors. Measurement of Error Standard error, mean squared error (variance), and mean absolute deviation. Mean absolute deviation (MAD): the average of the absolute value of the actual forecast error. Mean absolute percent error (MAPE): the average error measured as a percentage of average demand. MAPE = MAD / Average demand. Tracking signal: a measure of whether the forecast is keeping pace with any genuine upward or downward changes in demand. This is used to detect forecast bias. TS = RSFE/MAD RSFE: running sum of forecast errors. Casual Relationship Forecasting Casual relationship forecasting: forecasting using independent variables other than time to predict future demand. Like weather. Multiple Regression Analysis Qualitative Techniques in Forecasting Requires much expert judgement. Market Research used mostly for product research like ideas, likes and dislikes, and preferences. Panel Consensus panel of variety of people are more reliable forecast than narrower group. Historical Analogy Trying to forecast demand for a new product while using an existing product as a model. Like new coffee pots based on existing toasters. Like complementary, substitutable or competitive, and products as function of income. Delphi Method conceals identity of persons participating in study. WebBased Forecasting: Collaborative Planning Forecasting, And Replenishment (CPFR) Collaborative planning, forecasting, and replenishment (CPFR): an internet tool to coordinate forecasting, production, and purchasing in a firm's supply chain. Creation of a frontend partnership agreement, joint business planning, development of demand forecasts, sharing forecasts, inventory replenishment.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'