Chapter 12 Forecasting Forecasting Time Series Method Forecasting demand is a critical starting point in a business Forecast too high over production excess inventory high inventory cost waste Forecast too low under production inventory shortages late deliveries high shipping costs All forecasts are wrong so we have to try to be less wrong Forecast should respond to changes in demand but not overreact to unusual demand data Statistical techniques that use historical demand data to predict future demand Uses past data to forecast future demand Demand and sales are not the same Quantitative technique Not long term Relate the forecast to only one factor time Assume that what has occurred in the past will continue to occur in the future Includes moving average exponential smoothing adjusted exponential smoothing and linear trend line Regression Method Attempt to develop a mathematical relationship between demand and factors that cause its behavior Quantitative technique Linear regression o Mathematical technique that relates a dependent variable to an independent variable in the form of a linear equation Ex y a bx a intercept b slope of the line o o Correlation Qualitative Method A measure of the strength of the relationship between independent and dependent variables Use management judgment expertise and opinion to predict future demand Does not use past data long range Ex Delphi Method involves soliciting forecast about technological advances from experts Choice of Methods Depends On Time Frame o o Indicates how far into the future Short to mid range forecast Typically encompasses the immediate future Daily up to two years Long range forecast Usually a period of time longer than two years Demand Behavior o o o o o Trend Cycle A gradual long term up or down movement of demand Random Variations Movements in demand that do not follow a pattern An up and down repetitive movement in demand Seasonal Pattern Know it is going to happen Does not have to be the actual seasons winter fall summer spring Can be football season legislative An up and down repetitive movement in demand occurring periodically Causes of Behavior Forecasting Process Select a forecast and compute forecast for period of historical data Identify the purpose of forecast 1 2 Collect and plot historical data to identify patterns 3 4 Check forecast accuracy and make sure it is acceptable 5 Adjust forecast based on additional qualitative information 6 Monitor results and measure its accuracy along the way Moving Average Naive forecast Forecast is based on previous month s data o Demand in current period is used as next period s forecast o o Not a very good forecast o Ex January s demand is 120 February s forecast is 120 Simple Moving Average o Uses average demand for a fixed sequence of periods o o o o Stable demand with no pronounced behavioral patterns Sum the values in the period and divide by the length of the period Ex 3 month average Jan Feb Mar 3 Ex 5 month average Jan Feb Mar Apr May 5 Weighted moving average Adjusts moving average method to more closely reflect data fluctuations o Weights are assigned to most recent data o o Higher weight more responsive 1 3 o o Weights must between 0 100 must add up to 1 00 o o Sum of period weight period demand Ex 50 90 33 110 17 130 103 4 Lower weight less responsive 1 5 Exponential Smoothing Type of averaging method Reacts more to recent changes A way to assign weights to most recent data Weights most recent data more strongly Most and widely used accurate method Just like smoothing average it lags behind trends Seasonal Adjustments Repetitive increase decrease in demand Use seasonal factor to adjust forecast Forecast Accuracy Forecast Error o Difference between forecast and actual demand o o MAD Can be done for any time period Ex Jan Demand Jan Forecast Forecast Error o Mean absolute deviation o One of the most common MAPD o Mean absolute percent deviation o Measures how far in terms we are off Cumulative error E o E Sum of forecast error Average error bias Looks like MAD but no absolute value o o Not about accuracy it s about direction of error o o o Would like it close to 0 as possible o If bias is negative over forecast If bias is positive under forecast Ex E n Tracking Signal Monitors the forecast to see if it is biased high or low Tracking signal should always try to be close to 0 If it is 3 or more 3 or less change the forecast because it is not accurate Ex E MAD Chapter 14 Capacity Planning Capacity Maximum capability to produce How much we can do How much resources we have and how we use them Capacity Planning establishes overall level of productive resource for a firm Capacity increase depends on o o o Volume and certainty of anticipated demand Strategic objectives Costs of expansion and operation Best operating level is when of capacity utilization that minimizes unit costs Capacity Cushion of capacity held in reserve for unexpected occurrences Short Term Capacity Planning Work scheduling Disaggregation details Day to day operation Daily basis Medium Term Capacity Planning 9 months to 1 1 2 years Not intended for buildings and such Given our current resource how can we use it most effectively Aggregate Planning Also refers to sales and operations planning for product lines or families o o Determines resource capacity to meet demand over an intermediate time horizon o o Objectives Ex How many cars Ford makes in January Establish a company wide plan for allocating resources Develop an economic strategy for meeting demand Long Term Capacity Planning 3 basic strategies for timing of capacity expansion in relation to steady growth in demand lead lag and average Economies of Scale Fixed cost can be spread over a larger number of units Production operating costs do not increase linearly with output levels o Unit cost decreases as output volume increases o o o Quantity discounts are available for material purchase o Operating efficiency increases as workers gain experience Diseconomies of Scale o Unit cost increases as output volume increases Equipment overuse Worker fatigue Quality problems Reduced maintenance Rushed work Meeting Demand Strategies Adjusting Capacity Managing Demand o Proactive demand management Strategies for Managing Demand Resources to meet demand are acquired and maintained over the time horizon of the plan o o Minor variations in demand are handled with overtime or under time Offering products or services with counter cyclical demand patterns Ex Snow