University of San Carlos

School of Engineering

Engineering Management

________________________________

CASE STUDY
________________________________

Case Study 07: Forecasting

Submitted by:

Calilap, Woodrow Wilson

Crisologo, Sean Gabriel

Magdadaro, Dave Winkler

Yu, Justin Ray

SUBMITTED TO:
Engr. Leo Victor B. Amores, CIE, CLSSYB, Trained SSGB
Learning Assessment:

1. What are some of the potential benefits of a more formalized approach to forecasting?

• Stocks won’t be a problem as a formalized approach to forecasting will help you know
which products are more likely to run out of stock first and which products won’t sell more.
• Sales will be continuous and surges will be expected and easily handled because of the
forecasting data.
• Better control of the inventory and the ordering of new stocks for the company.
• Better decision making made available by the data from the forecasting.
• Forecasting data can be used to plan for future expansions of the company.

2. Prepare a weekly forecast for the next four weeks for each product. Briefly explain why you
chose the methods you used. (Hint: For product 2, a simple approach, possibly some sort of naive/
intuitive approach, would be preferable to a technical approach in view of the manager’s disdain
of more technical methods.)

Week Product - 1 Product - 2

1 50 40

2 54 38

3 57 41

4 60 46

5 64 42

6 67 41

7 90* 41

8 76 47

9 79 42

10 82 43

11 85 42

12 87 49

13 92 43

14 96 44
For Product 1
As seen from the graph the trend of product 1 is rising except in the 7th week where it has
recorded a product spike of 90 units and goes down again in the 8th week to 76 units. In order to
correctly interpret the trend, we would use a linear regression since the graph indicates a strong
linear relationship between the variables. Performing the linear regression:

Product - 1
Week (X) (Y) ⅀XY ⅀Y^2 ⅀X^2

1 50 50 2500 1

2 54 108 2916 4

3 57 171 3249 9

4 60 240 3600 16

5 64 320 4096 25

6 67 402 4489 36

7 90 630 8100 49
 8 76 608 5776 64

9 79 711 6241 81

10 82 820 6724 100

11 85 935 7225 121

12 87 1044 7569 144

13 92 1196 8464 169

14 96 1344 9216 196

105 1039 8579 80165 1015

(𝑛⅀𝑋𝑌) − (⅀𝑋)(⅀𝑌)
𝑏 =
(𝑛⅀𝑋 2 ) − (𝑛⅀𝑋)2

(14 ∙ 8579) − (105)(1039)

𝑏 =
(14 ∙ 1015) − (105)2

11,011
𝑏 = = 3.45
3185

𝑎 = ȳ − 𝑏𝑥̄
1039 105
𝑎= ( ) − 3.45𝑥( ) = 48.34
14 14
𝑌 = 𝑎 + 𝑏𝑋
𝒀 = 𝟒𝟖. 𝟑𝟒 + 𝟑. 𝟒𝟓𝒙

Forecast for 4 weeks (weeks 15-18)

𝐹15 = 48.34 + 3.45(15)
𝐹15 = 100.09
𝐹16 = 48.34 + 3.45(16)
 𝐹16 = 103.54
𝐹17 = 48.34 + 3.45(17)
𝐹17 = 106.99
𝐹18 = 48.34 + 3.45(18)
𝑭𝟏𝟖 = 𝟏𝟏𝟎. 𝟒𝟒

week forecast

15 100.09

16 103.54

17 106.99

18 110.44

Reason for performing the linear regression:

When plotting the raw data in a graph it shows a strong positive linear relationship among
the variables. By getting the linear regression, the forecast of the next 4 weeks would be easily
predicted since the data has already shown a positive linear trend.

For product 2
As seen from the graph, product 2 has a much more complicated trend compared to product
1. It has a much-complicated trend since the values are closer to each other and does not exhibit a
positive or negative trend. It is observed from the graph that every 4 weeks there is a spike from
the demand. For week 16 the observed forecast would be 52 since the trend for the past weeks the
demand would increase 1 unit every week spike since the previous week the increase was 2 units
it is assumed that the increase for the week 16 is 3 units. In non-spike weeks the trend is that the
units will increase 1 unit per week. For week 15 the moving average is taken from weeks 9,10,11
before the spike occurred. The forecast would be (42+43+42)/3 would be now equal to 42.33 plus
1 since it is a non-spike week. For week 15 the forecast is 43.33. For the weeks 17 and 18 the
moving average is taken from weeks 13, 14, 15 that would be (43+44+43.33)/3 would now be
equal to 43.44 plus 1 since it is a non-spike week. For week 17 and 18 the forecast is 44.44.
 Week Forecast

15 43.33

16 52

17 44.44

18 44.44

Reason for performing the intuitive approach:

Intuitive approach was used to predict the data since from the graph the data shows a
complex pattern. This approach is also done so that the forecasts and planning would be
improved and the stockouts would be also decreased.