Week 2 Assignment

Amit Ramsahay Verma (2219877)

University Canada West

OPMT 620- Operations Management

Aida Esmaeilidouki

1
 Questions:

Q-1: Please respond to the weekly forum 2 provided in Week 2.

A-1: Done.

Q-2: Forecast demands for using weighted moving average. Please apply weights of 0.3
for most recent period, then 0.4, then 0.3
A-2: Weights = 0.3, 0.4, 0.3

F7= (0.3)7-1(25)7-1+ (0.4)7-2(22)7-3+(0.3)7-3(18)7-3

F7= 0.3(25) + 0.4(22) +0.3(18) = 21.7
F6=0.3(22) +0.4(18) +0.3(14) = 18
F5=0.3(18) +0.4(14) +0.3(12) =14.6
F4=0.3(14) +0.4(12) +0.3(6) =10.8

Period Demand Forecast

1 6
2 12
3 14
4 18 10.8
5 22 14.6
6 25 18
7 28 21.7
8 25.0

Q-3: Given the following data:

Prepare a forecast using smoothing constant  = 0.40.

a) Subjective estimate (for this example, use 60).
b) First actual value (naïve approach).
A-3: a) The forecast F1 is calculated as 60 ( By subjective Approach)

Ft = Ft-1 +  (At-1 – Ft-1)

Ft = forecast value for period t, At = demand at period t, α = smoothing constant (0≤ α≤1)

2
 F2=F1+0.40(A1-F1) = 60+0.40(63-60) = 61.2
F3=F2+0.40(A2-F2) = 61.2+0.40(52-61.2) = 57.52
F4=F3+0.40(A3-F3) = 57.52+0.40(56-57.52) = 56.912

Period Actual Forecast Calculations

1 63 60
2 52 61.2 60+0.40(63-60)
3 56 57.52 61.2+0.40(52-61.2)
4 62 56.912 57.52+0.40(56-57.52)

4) Given the following data:

c) What are the exponential smoothing forecasts for periods 2-5 using  =0.7?
d) Use naïve approach for 1st week.

Ft = Ft-1 +  (At-1 – Ft-1)

Ft = forecast value for period t, At = demand at period t, α = smoothing constant (0≤ α≤1)

F2= 830+0.7(830-830) = 830

F3=830+0.7(765-830) = 784.5
F4= 784.5+0.7(630-784.5) = 676.35
F5= 676.35+0.7(640-676.35) =650.905

Week Demands 0.7

1 830 830
2 765 830
3 630 784.5
4 640 676.35
5 650.905

3
 Q-5): Predict quarterly demand for a certain loveseat. The series has both trend and
seasonality andthe Quarterly relatives are: Q1= 1.20, Q2 = 1.10, Q3 = 0.75, Q4 =0.95. If the
trend equation equals to yt=124+7.5t (t = quarter 15), Predict demand for quarter 2
(where t 15 = quarter 2).
A-5): Q1= 1.20, Q2 = 1.10, Q3 = 0.75, Q4 =0.95
yt=124+7.5t (t = quarter 15)
for quarter 2 of ->t = 15

Y15= 124 + 7.5×15 = 236.5

F15 = 236.5×1.10 = 260.15

Q-6): Given the following data, compute MAD, MSE, and MAPE for the following data.
Then, please compute the standard deviation of forecast error using the first eight
months.
A-6):

MAD (Mean Absolute Deviation)

MSE (Mean Squared Error)

MAPE (Mean Absolute Percent error)

Demand Forecast (A-F) Forecast |Error|/Actual

Months Forecast
(Actual) Error (|Error| (Error)^2 *100
1 210 222 -12 12 144 5.71%
2 240 235 5 5 25 2.08%
3 230 225 5 5 25 2.17%
4 240 245 -5 5 25 2.08%
5 280 265 15 15 225 5.36%
6 222 250 -28 28 784 12.61%
7 220 220 0 0 0 0.00%
8 250 273 -23 23 529 9.20%
1892 93 1757 39.22%

4
 MAD= 93/8 MSE= 1757/8 MSAPE= 39.22%/8

MAD 11.625
MSE 219.625
MAPE 4.90%

The standard deviation of forecast error using first eight months is 9.8262= 9.83.

Q-7). Below are a pharmacy’s actual sales and forecasted demand for a certain
prescription drug for 5 months. How accurate is their forecast? Calculate MAD and MSE
and create a control chart.

A-7)
To find the accuracy of the forecast we should know the value of MAPE (Mean Absolute
Percent Error) Accuracy= 100%-MAPE

Period (n)= 4

Forecasted Forecast (A-F) Forecast |Error|/Actual

Actual sales
demand Error (|Error| (Error)^2 *100
1 215 n/a
2 235 255 -20 20 400 8.51%
3 200 203 -3 3 9 1.50%
4 305 325 -20 20 400 6.56%
5 310 320 -10 10 100 3.23%
1265 53 909 19.79%

MAD (Mean Absolute Deviation)

MSE (Mean Squared Error)

MAPE (Mean Absolute Percent error)

5
MAD= 53/4 MSE= 909/4 MAPE= 19.79%/4

MAD 13.25
MSE 227.25
4.95%
MAPE

So, the Accuracy of the forecast = 100%-MAPE = 95.05%

The control limits are usually chosen as a multiple of the standard deviation of forecast errors.
S=√𝑀𝑆𝐸= 15.075
limits of 0 ± 2s (i.e., O ± 2 standard deviations).
This is because the average of forecast errors is 0.

UCL: 0+ z√MSE= 0+ 2√227.25= 30.15

LCL: 0 - z√MSE=0-2√227.25= -30.15
The control limits are -30.15+30.15
As we can see that all the errors are within the control limit.

Month Forecast Error Avg UCL LCL

1
2 -20 -13.25 30.15 -30.15
3 -3 -13.25 30.15 -30.15
4 -20 -13.25 30.15 -30.15
5 -10 -13.25 30.15 -30.15

6
8) Freight car loadings during a 14-week period at a port are:
a. Compute a linear trend line for freight car loadings using of Excel’s Trendline,
with displayEquation on chart option.
b. Use the trend equation to predict loadings for weeks 15 and 16.

A-8)-a)

Week Number Week Number

1 232 8 385
2 250 9 420
3 270 10 465
4 262 11 485
5 320 12 498
6 352 13 520
7 375 14 555

By plotting the figures, we get a Linear trendline with equation as Y=25.76x+ 191.73

b)
Now if we want the values of week 15 and 16, we use the equation Y=25.76x+ 191.73

7
 For Week 15
X=15
Y=25.76(15) + 191.73= 578.13

For week 16
X= 16
Y= 25.76(16) + 191.73= 603.89

For Week 15=Forecasted Loading= 578.13

For week 16= Forecasted Loading= 603.8

Q-9)-Use of simpler linear regression analysis assumes that:

a. Variations around the line are non-random.
b. A straight line will be determined that maximizes the sum of deviations of the data
points.
c. Predictions are to be made only within the range of observed values of the
predictorvariable.
d. Deviations around the line are not normally distributed.
e. Predictions can be made outside the range of observed values of the predictor
variable.

A-9)- “c” Predictions are to be made only within the range of observed values of the
predictor variable.
Simpler linear regression analysis presupposes that predictions are limited to the predictor
variable's observed range of values. Extrapolating beyond this range may result in inaccurate
predictions since the relationship between the variables may not hold true outside of the observed
data range.

Q-10)-Please answer the following True or False questions.

a. When error values fall outside the limits of a control chart, this signals a need for
correctiveaction. (True/False)
b. When all errors plotted on a control chart are either all positive, or all negative,
this showsthat the forecasting technique is performing adequately. (True/False)
c. A random pattern of errors within the limits of a control chart signals a need
for correctiveaction. (True/False)

A-10)- a. True

8
When error numbers exceed the bounds of a control chart, it indicates that the process is out of
control and that corrective action is required to explore and address the sources of variation.

b. False
When all the mistakes plotted on a control chart are either all positive or all negative, it does not
always indicate that the forecasting technique is working properly. A constant bias in one way
suggests a systemic inaccuracy, which could indicate a fault with the forecasting model or data.

c. False
A well-performing forecasting technique should have a random pattern of mistakes within the bounds
of a control chart. The presence of randomness in mistakes indicates that the forecasting model is
capturing most of the variation, and any remaining errors are attributable to random fluctuations
inherent in any forecasting process.

9
1
0