PM & Forecasting

Question 1

When to use Earned Value Analysis

Answer 1

1. To compute how far ahead or
   behind schedule a project is.
2. To compute how over or under
   budget a project is.

Question 2

Budgeted Cost of Work Scheduled
BCWS

Answer 2

% Schedulled x Budget

BCWS is the amount of money you planned to spend on the project up to a certain point in time based on the scheduled tasks.

Question 3

Eg. Budgeted Cost of Work Scheduled BCWS

Answer 3

Example:
Total project budget = $100,000

By Week 4, you planned to complete 40% of the work.

Then:
𝐵𝐶𝑊𝑆 = 40%×$100,000 =$40,000

Question 4

BCWC Budgeted Cost of Work Completed to date for the activity

Answer 4

% completed x Budget

How much we expected to spend based on the actual amount of completed work

Question 5

Total BCWS

Answer 5

Sum of BCWS for each activity

Question 6

Total BCWC (Earned Value)

Answer 6

Sum of BCWC for each activity

Question 7

How do you calculate how far ahead or behind schedule an ACTIVITY is?

Answer 7

% Schedulled - % Completed / Schedulled x 100

OR

BCWS - BCWC / BCWS x 100 %

Question 8

How do you calculate how far ahead or behind schedule the PROJECT is?

Answer 8

(BCWS total - BCWC total / BCWS total) X 100

Question 9

ACWC

Answer 9

Actual Cost of Work Completed

Question 10

Actual Cost of Work Completed

Answer 10

How much we actually spend based on the actual amount of completed work

Question 11

Forecasting

Answer 11

Statistical estimate of future demand, that can be used to plan current activities.

Question 12

Qualitative Forecasting Methods

Answer 12

1. Executive Judgement
2. Market Research
3. Panel Concensus

Question 13

Quantitative Forecasting Methods

Answer 13

Typical Time Series of Past Demands

Question 14

Forecastng Notation

Answer 14

At and Ft

Question 15

At

Answer 15

The actual demand at time t

Question 16

Ft

Answer 16

The forecasted demand at time t

Question 17

Methods to forecast demand at time t Ft

Answer 17

1. Simple Moving Averages
2. Weighted Moving Averages
3. Exponential Smoothing

Question 18

n-period Simple Moving Averages

Answer 18

Ft = At1+At2+…+At-1 / nt-1

Question 19

n-period weighted moving average (WMA)

Answer 19

Ft = w1xAt1 + W2xAt2 +…+ wnxAt-1

W1 is the most recent observed demand, greater weight. De abajo para arriba

Can assign different weight to different time periods

Question 20

Forecast error at time t

Answer 20

Ft - At

Question 21

How to evaluate a forecast

Answer 21

Using the mean absolute error (MAE)

Question 22

Mean Absolute Error - MAE

Answer 22

∑ ​∣ Forecast t − Actual t ∣ / T

Question 23

Advantages of Forecasting with more periods

Answer 23

More data = less randomness and more reliable averages

If there’s no trend, more data = lower forecasting error.

Question 24

What are the downsides of using too many periods in a forecast?

Answer 24

Slower to react to real changes in demand

If there’s a trend, more data can actually make the forecast worse

Question 25

So… how many periods should I use in forecasting?

Answer 25

If demand is stable ➜ Use more periods to smooth out noise If demand is changing (trending) ➜ Use fewer periods to react faster

Question 26

Simple Moving Average (SMA)

Answer 26

Is a special case of a weighted moving average (WMA) where all weights are equal. Each period is given the same weight

Question 27

Exponential Smoothing

Answer 27

Weighted moving averages consider all times periods. It gives more weight to recent data while still considering all past periods — unlike moving averages which only look at the last n periods.

Question 28

Smoothing Factor

Answer 28

α (0 < α < 1)

Question 29

Exponential Smoorhing (ES)

Answer 29

Ft = α⋅At−1+ (1−α)⋅Ft−1 ​

Question 30

Ft

Answer 30

Forecast for current period

Question 31

Ft-1

Answer 31

Forecast from previous period

Question 32

At-1

Answer 32

Actual value from the previous period

Question 33

What is exponential smoothing used for in forecasting?

Answer 33

To give more importance to recent data while still considering all past data.

Question 34

Higher α (e.g., 0.8)

Answer 34

More weight to recent data → reacts quickly to changes

Question 35

Lower α (e.g., 0.2)

Answer 35

More weight to older data → smoother, but slower response

Question 36

F1

Answer 36

Inital Forecast Value

Question 37

First Period Rule

Answer 37

For the First Period the F1=At-1

Question 38

3 Meassures of forecasts are used

Answer 38

1. Mean Absolute Error (MAE) 2. Mean Squared Error (MSE) 3. Mean Absolute Percent Error (MAPE)

Question 39

e

Answer 39

Error

Question 40

e formula

Answer 40

Actual - Forecast At - Ft

Question 41

Mean Absolute Deviation (MAD)

Answer 41

Measures the average size of forecasting errors — ignoring whether they’re positive or negative. Tells you, on average, how far off your forecasts are from the actual values.

Question 42

Mean Absolute Deviation (MAD) FORMULA

Answer 42

∑∣Actual - Forecast∣ / n

Question 43

Mean Squared Error (MSE)

Answer 43

∑ (Actual - Forecast )2 / n

Question 44

Mean Absolute Percent Error (MAPE)

Answer 44

∑∣ ∣ Actual - Forecast ∣ / Actual ∣ x 100 / n