Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

operations management > 2 - forecasting > Flashcards

2 - forecasting Flashcards

1
Q

what type of demand requires forecasting and which one doesn’t ?

A
  • independent demand (finished goods and spare parts) requires forecasting
  • dependent demand (raw materials and components) does not require forecasting
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

what are the characteristics of forecasting?

A
  • based on the assumption that past will repeat itself
  • seldom perfect: almost always wrong
  • family and aggregated product forecast is more accurate than individual product forecast
  • longer forecast period = less accurate
  • volatile demand = difficult to make accurate forecasts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

what internal factors influence demand?

A
  • price lvls
  • promotion and ads
  • changes to product line
  • intro of substitute products
  • new points of sales
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what external factors influence demand? (PESTEL)

A
  • Economic context (inflation rate, interest rate, unemployment rate, etc.)
  • Seasonality
  • Competition’s actions (new entries in the market)
  • Product’s life cycle (declining stage, plateau stage?)
  • Customer preferences
  • Government regulations (taxes, quotas, security regulations, etc.)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what are the demand components?

A
  • random or residual (any other error/effect)
  • trend (increasing or decreasing)
  • seasonality (increase/decrease regularly and predictably over a period)
  • life cycle (bigger interval than seasonality)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what are the 2 forecasting methods? and which one is for short/long term?

A
  • quantitative methods : time series (short term)

- qualitative methods : opinions and historical analogies (long term)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

what are the 4 qualitative methods for forecasting?

A
  • expert panel
  • delphi method
  • consumer market survey
  • historical analogies
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

when should qualitative methods be used ?

A
  • new products
  • historical data unreliable
  • to validate quantitative model
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

what are the advantages of qualitative methods?

A
  • Take intangible factors into consideration (that quantitative methods don’t).
  • Useful when there are little data available (new product, new market, new business unit).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what are the disadvantages of qualitative methods?

A
  • Long consultation process
  • High risk of getting a biased forecast
  • Expensive
  • Usually not precise
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what are the quantitative methods and what are they based on?

A

quantitative methods are based on historical data or on associations between different variables

  • time series (monthly sales of previous periods)
  • associative models
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

what are the advantages of quantitative methods?

A
  • Easy to use once the right model has been developed.
  • Data collection is quick and easy, since most of the required information is already in the information systems of the company, (ex. previous sales) or readily available (ex. consumer price index).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

what are the disadvantages of quantitative methods ?

A

-Do not take new factors into consideration

“ It’s like driving a car by looking in the rear-view mirror.”

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

what are the characteristics of the time series quantitative method?

A
  • assume that was has occurred in the past will continue to occur in the future
  • relate the forecast to only one factor: time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

what are the 4 time series methods?

A
  • naive forecast
  • simple moving average
  • weighted moving average
  • exponential smoothing
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

what is naive forecast?/

A

A naive forecast for one period is equal to the demand value of the previous period (demand last week was 250, so this week’s demand is 250)

17
Q

when can we use naive forecast?

A

used when presence of a strong seasonality (no trends ou random variations)

18
Q

what assumptions must be made when using simple moving average of k periods?

A
  • no trend, seasonality or cyclical component

- equal weight to last k observations

19
Q

what should be considered when choosing the size of k for the simple moving average?

A
  • if k is small: forecasts will react rapidly to real changes (trends) but could also react to noise (random variations)
  • if k is large: less affected by random components and forecasts will be slow to react to real changes = gives more moderate and stable forecast
20
Q

what is the principal of the weighted moving average?

A

Weighted moving average places different weight on the observed periods.Usually, greater weight will be given to the most recent data.

21
Q

what is different from the simple moving average and the weighted moving average?

A

reflects upward or downward trends more quickly, while smoothing out the effect of random variations better than a smaller k value would

22
Q

what are some difficulties of the weighted moving average?

A
  • need to keep a lot of data

- need to assign weights (arbitrary)

23
Q

what is the principal of exponential smoothing?

A

It’s a weighted moving average but more refined and easier where we determine only one weight (α), the importance given to the previous period (using only one weight for the best data). The rest of the weight (1-α) goes to the previous period’s forecast.

24
Q

what are the assumptions to make when using exponential smoothing?

A
  • no trend, seasonality or cyclical component

- the most recent data usually receive a higher weight

25
Q

what must be considered when choosing a small a in exponential smoothing?

A

If α is small: noise caused by random demand is reduced, but forecasts will be slow to react to real variations.

26
Q

in a context where there is a lot of variation in demand, it is better to choose a smaller or a bigger smoothing constant?

A

when there is a lot of random variation in demand, it is better to choose a smaller smoothing constant (a)

27
Q

what must be considered when choosing a large a in exponential smoothing?

A

If α is large: Forecasts will react rapidly to real variations, fluctuations in demand but could react to «noise»: this is an advantage if the variations reflect real trends, but if they are merely noise caused by random effect, it is a disadvantage

28
Q

when can a large smoothing constant be used?

A

when the random component has little influence on demand, a higher smoothing constant can be used

29
Q

the higher the constant a, the higher…

A

the greater the importance given to the deviation between actual demand and the forecast

30
Q

what are the 3 common measures of error?

A
  • individual error
  • mean error
  • mean squared error (MSE)
31
Q

how do you calculate individual error and what are the 2 types of errors linked to it?

A

ei = Di-Fi

  • bias
  • random variations
32
Q

what is the mean error and how do you calculate?

A

it is the average of all forecasting errors in view of the actual demand
= (Di-Fi)/n

33
Q

what does the mean error identifies?

A

it allows the identification of a systematic bias in the forecasting model

34
Q

the mean error should converge to…

A

the mean error should converge to zero

35
Q

what does the mean squared error amplifies and how is it calculated?

A

the mean squared error amplifies the impact of the larger errors (large deviations have much greater weight)
MSE = (D-F)^2/n

operations management (4 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions