Quiz 2 (Sessions 9-12 excl. Fixed-Period or Periodic (P) system and excel) Flashcards
forecasting
process of predicting a future event
Underlying basis of all business decisions
For example:
– Production
– Inventory
– Personnel
– Facilities
qualitative forecasting
typically used when the situation is vague and little data exits
-eg, new products, new tech
-involves intuition and experience
quantitative methods
Typically used when the situation is ‘stable’ & historical data exists
-eg, existing products, current tech
-involves mathematical equations
types of quantitative approach
time series forecasting
-naive approach
-moving average, weighted moving average
-exponential smoothing
-(trend projection)
-multiplicative seasonal model
(associative forecasting methods)
-(linear regression)
-(correlation analysis)
time series
-set of evenly spaced numerical data
–obtained by observing the variable of interest at a regular time periods
-use data from previous periods to predict results in future periods
-most appropriate (accurate) when near-future periods results are influenced by similar conditions as those in the past
naive approach
-assumed demand in the next period us the same as demand in the most recent period
-eg, if may sales were 48, then June sales will be 48
-Ft = At-1 -> forecast in time t = actual demand in time t-1
-often acts as a starting point that we can use as a comparison to other forecasting approaches
-primarily used for its efficiency
moving average method
-MA is series of arithmetic means
-useful when we can assume that market demands will stay fairly steady over time; tends to smooth out short term irregularities in the data series
-less reactive than the naive approach to random changes in demand; using more periods reduced reactiveness
WMA
-often used to make the moving average forecast more reactive
–ie, older data is considered less important so has less weight
exponential smoothing method
-special form of weighted moving average
–weights decline exponentially; heaviest weight is most recent
-reqs smoothing constant (alpha)
–ranges from 0 to 1
-involves little record keeping of past data
-know rewritten formula
smoothing constant
-between 0 and 1; typically between 0.05 and 0.50 for business
–alpha closer to 0: heavily weights past periods (less reactive)
–closer to 1: heavily weights recent data - highly reactive
-evaluating forecasting error helps to select the most effective alpha value
forecasting error equation
forecast error = actual - forecast
-also know mean absolute deviation (MAD) equation
multiplicative seasonal model (steps to develop seasonal forecast)
- find avg historical demand for each season
- compute avg teman over all seasons
- seasonal index for each season
- estimate next period’s total demand (eg, next years annual demand)
- divide this estimate by the number of seasons and then multiply the resulting value by each seasonal index to provide seasonal forecasts
inventory types
- raw materials (RM)
- work in process (WIP)
- finished goods inventory (FG)
- maintenance/repaid/operating supplies (MRO)
IM decisions
-what, how much, when to order
- how to monitor
IC: holding costs
VC that increase proportionally with volume held
IC: ordering costs
associated w costs of placing orders and receiving goods
IC: setup costs
cost to prepare a machine or process for producing an order
inventory expense
-costs incl shrinkage, obsolescence, interest payments for working capital, insurance, storage-space rental etc that increase proportionally w volume and value of inventory items
-firms can only est. holding cost rates; typically set annually, generally between 20-30% of the items’s value, calculated on the avg volume held
-rates may be higher where inventory is perishable, subject to rapid obsolescence, or prone to shrinkage
hold down avg volume
for same annual inventory volume buying frequently in small lots reduces age inventory level
reduce annual ordering costs
buying many small lots means more ordering costs annually - may offset holding cost savings