Chapter 12 Forecasting 03 04 2013 Forecast prediction of what will occur in the future Accurate forecasting determines how much inventory a company must keep at various point along its supply chain Methods subjective based on mathematical formulas statistical techniques that use historical demand data to predict a Qualitative b Quantitative c Time series future demand demand and factors that cause its behavior d Regression methods attempt to develop a mathematical relationship between Choose your method based on a Time frame how far into the future is the forecast i Short to mid range immediate future daily up to 2 years ii Long range longer than 2 years b Demand behavior i Trend gradual long term up or down movement of demand ii Random variation movements in demand that do not follow a pattern iii Cycle up and down repetitive movement in demand iv Seasonal an up and down repetitive movement in demand occurring periodically c Causes of behavior Forecasting process Identify the purpose of forecasting 1 2 Collect historical data 3 Plot data and identify patterns 4 Select a forecast model that seems appropriate for data 5 Develop compute forecast for period of historical data 6 Check forecast accuracy with one or more measures 7 Is accuracy of forecast acceptable 8 a b if yes forecast over planning horizon if no select new forecast model or adjust existing parameters 9 Adjust forecast based on additional qualitative information 10 Monitor results and measure forecast accuracy Timesheets methods uses historical demand data overtime to predict future demand Only factor is time good for short term forecasting A Moving average sometimes called na ve or intuitive demand in current period is used as next periods forecast i Simple moving average past to develop a forecast helps smooth out uses several demand values during the recent n cid 229 Di i 1 n MAn N number of period in moving average Di demand in period i Good for stable demand with no pronounced behavior patterns Longer period moving averages react more slowly to recent demand changed than shorter period moving averages Shorter period moving averages are more susceptible to simple random variation ii Weighted moving average closely related data fluctuations adjusts moving average method to more Wi weight of period i between 1 100 WMA n DW i i n 1 i Weights are assigned to the most recent data If most recent period are weighted too heavily the forecast might overreact to a random fluctuation If weighted too lightly the forecast might underreact to actual changed in demand behavior B Exponential Smoothing an averaging method that reacts more strongly to recent changes in demand cid 229 F t 1 forecast for next period Dt actual demand in the present period Ft the previously determined forecast for present period a weighted factor referred to as the smoothing constant F t 1 a D t 1 a F t Smoothing constant weighting factor given to the most recent data in exponential smoothing between 0 1 the closer it is to 1 the greater the reaction to the most recent term if it is 0 the forecast does not reflect the most recent data at all i adjusted exponential smoothing trend adjustment added T and exponentially smoothed trend factor Tt last periods trend factors B smoothing constant for trend between 0 1 AF t 1 F t 1 T t 1 b T t 1 F t 1 F t 1 b T t C Season adjustments repetitive increase decrease in demand adjusts for seasonality by multiplying the normal forecast by a seasonal factor Forecast Accuracy Forecast error difference between forecast and actual demand MAD mean absolute deviation the absolute difference between the forecast average and demand t time period Dt demand in period t Ft forecast for period t N of periods o the lover the value of MAD relative to the magnitude of data the more A MAPD mean absolute percent deviation the absolute error as a percentage od MAD FD t t n accurate the forecast demand rather than the period FD t t MAPD D t cid 229 Bias t e n TS t FD t MAD cid 229 E B Cumulative Data sum of the forecast errors te FDe t t t If E is positive forecast probably constantly lower than actual demand or biased low If E is negative forecast is probably constantly higher than actual demand or biased high C Average Error per period average of cumulative error Forecast Control monitor forecast error overtime to make sure that the forecast is performing correctly Tracking Signal monitors forecast to see if it is biased high or low As long as tracking signal is within control limits the forecast is in control control limits usually 2 to 5 MAD 1MAD 8 cid 229 cid 229 cid 229 cid 229 Chapter 7 Capacity Facilities Design 03 04 2013 Capacity maximum capability to produce Capacity Planning establishes the overall level of productive recourses for a firm Affect product lead times customer responsiveness operation costs and a firms ability to compete A Capacity lead strategy B Average capacity strategy capacity is expanded to coincide with average expected capacity is expanded in anticipation of demand growth C Capacity lag strategy capacity is increased after an increase in demand has been demand discovered How much to increase capacity depends on 1 Volume and certainty of anticipated demand 2 Strategic objectives in terms of growth customer service and competition 3 Cost of expansion and operation an alternative to expanding capacity is outsourcing Best operating level percent of capacity that minimizes unit cost Capacity cushion percent of capacity held in reverse for unexpected occurrences Diseconomies of scale when higher levels of output cost more per unity to produce Economies of scale when it costs less per unit to produce high levels of output Chapter 14 Sales Operations Planning 03 04 2013 Operations Planning aggregate planning process that determines the resource capacity a firm will need to meet its demand Aggregate plans are developed for product lines or product families not individual products Goals Establish a company wide plan for allocating resources Develop an economic strategy for meeting demand Strategies for Managing Demand o Shifting demand into other times periods to smooth out fluctuations ex incentives sales promotions advertising campaigns o Offering products services with counter cyclical demand patterns mountain biking in the summer on ski slopes B Chase Demand Strategies for Adjusting Capacity A Level production meet needs hiring firing C Peak demand D Overtime under time fluctuations are not extreme E