Decision Support System Assignment - Analysis, Solutions and Results

This assignment delves into various aspects of Decision Support Systems (DSS). It begins by examining decision-making under different conditions, including certainty, risk, and uncertainty, and provides analyses of investment scenarios. The assignment then applies Monte Carlo simulation to determine average monthly profit based on given data, followed by a discussion of the impact of changing variables. Furthermore, it uses the high-low method and regression analysis to evaluate cost behavior and model overhead costs, comparing different regression models and assessing their significance. Finally, it explores break-even analysis to determine the number of units required to reach the break-even point for two products. The assignment provides detailed calculations, model comparisons, and recommendations, making it a comprehensive study of DSS applications.

DECISION SUPPORT SYSTEM
[Pick the date]
Student id and name

Question 1
(a) The decision making under different situations has significant differences outlined below.
 “Decision making under certainty”- In this, a host of reliable information is available in
relation to the future and the present and hence the decision making is neither challenging,
nor time consuming. Also, considering that these are frequently made decisions, hence clear
guidelines or processes are available to take decision in such situations (Holsapple &
Whinston, 2013).
 “Decision making under risk” - In this, the information is limited in scope but based on the
same various alternatives are clear and also the subjective probabilities can be ascertained
with reasonable accuracy. In relation to tools, mathematical models are frequently used
which mirror the future possibilities in terms of monetary values based on inputs from
decision makers thus aiding decision making (Medhi, 2001).
 “Decision making under complete uncertainty” – In this, no reliable information is available
with regards to the likely outcomes and their respective probabilities. These decisions are
exceptionally difficult to rake and time consuming as the approach becomes very critical in
the lack of other tools to rely on. Creativity in outlook and approach is required for decision
making. However, these decisions are quite infrequent.
(b) Results
(1) Optimist would chose “share market” because it provide maximum profit as shown in
the table.
(2) Optimist would chose “bonds” as shown in the table.
(3) “Real Estate” would be as per the impact of criteria of regret as shown in the table.
1 | P a g e

(4) Expecte d monetary value for all the three investment option which are represented in the
table highlighted below:
After analyzing the above table and computation, it can be said that “bonds investment” has
maximum value of expected monetary value and hence, it should be chosen for investment.
(5) EVPI “Expected value of perfect information” is computed as shown below:
2 | P a g e

Question 2
(a) Jerry has two options of shop i.e. large shop and small shop to start his bicycle business.
The selection of best suitable option between these two would be mainly depends on the
profit value. Shop with high profit would be considered.
Large shop is provided high profit ($20,000) for Jerry and hence, he should select large shop to
start his bicycle business.
(b) For good and poor market the prior probability revision table is shown below:
Favorable study Prior probability revision
Unfavorable study Prior probability revision
3 | P a g e

(c) Posterior probability for unfavorable study is furnished below:
(d) The objective is to compute whether the market research conducted by Jerry’s friend
would result gains or loss.
Favorable study
4 | P a g e

Unfavorable study
Computation of expected gains/losses
5 | P a g e

Question 3
(a) Monte Carlo simulation to find the avg. monthly profit based on the given data (Holsapple &
Whinston, 2013).
Normal view:
Formula view of model
6 | P a g e

(a) “Avg. monthly profit” = $561.39
(b) The new model is formed by taking the new avg. sale price range from ($80) to ($100)
and the profit margin range from (22%) to (32%).
7 | P a g e

8 | P a g e

Date: 22 September, 2017
Dear Sir
In line with the instructions, suitable amendments have been made in the simulation model to
reflect the higher pricing. This has given a boost to the profits generated as the product
profitability has increased without having any adverse effect on demand. But, in relation to
implementation of this suggestion, a cautionary approach is recommended considering that
increased price may lead to sales volume dampening. As a result, even though profit margins
may increase but the overall profits would not improve to the extent predicted in the model.
Further, it could have larger strategic implications which need to be explored further. Also,
during implementation monitoring of sales on a frequent basis would also be desirable so that
any potential issues may be dealt in a prompt manner with minimal or no damage whatsoever.
Yours faithfully
STUDENT NAME
9 | P a g e

Question 4
(a) High- low method
Required parameters:
 Variable cost per unit produced
 Fixed cost
 Dependent variable
 Independent variable
Hence,
“Variable cost per unit produced” ¿(48000−46000)/(3800−1800)=$ 1
Fixed cost ¿ 48000−( 3800× 1)=$ 44,200
Dependent variable ¿
−head cost (OH )
Independent variable ¿ Machine hours ( MH )
Regression equation by using high- low method
Machine hours is given as 3000.
Hence,
¿−head cost ( OH )=44200+ ( 1× 3000 )
¿−head cost ( OH ) =$ 47,200
(b) Regression analysis
Model 1
10 | P a g e

Hypothesis testing
The value of t stat -0.30
Corresponding p value 0.77
Significance level 5%
Decision P > Significance level
Fails to reject null hypothesis and hence, the “slope is not significant and model is not good.”
11 | P a g e

Model 2
Hypothesis testing
The value of t stat 6.28
Corresponding p value 0.00
Significance level 5%
Decision P < Significance level
Reject null hypothesis and hence, the “slope is significant and model is good.”
12 | P a g e

Model 3
Hypothesis testing
For machine hours
The value of t stat -0.27
Corresponding p value 0.79
Significance level 5%
13 | P a g e

Decision P > Significance level
Fails to null hypothesis and hence, the “slope is not significant and model is not good.”
Hypothesis testing
For batches
The value of t stat 5.87
Corresponding p value 0
Significance level 5%
Decision P < Significance level
Reject null hypothesis and hence, the “slope is significant and model is good.”
(c) The key characteristic for the regression model should be the predictive power and the ability
to account for the changes in the dependent variable i.e. overhead cost. This is reflected through
the use of coefficient of determination or R2. But deployment of the same for decision making
would be inappropriate here as there are two simple regression models while one multiple
regression model. To account for the difference in the independent variables present, adjusted R2
use is recommended (Lind, Marchal & Wathen,2012).
Based on this, the regression model that is based on batches only as the predictor variable is the
preferred choice.
14 | P a g e

(d) Best regression model – 2
Hence,
Batches = 150
Question 5
(a) Computation of unit contribution margin (Lind, Marchal & Wathen,2012)
(a) Units of product B at Break- even point (total cost = total revenue)
(b) Units of product A at Break- even point (total cost = total revenue)
15 | P a g e

(c) (i) Number of units of product A and product B
(ii) Number of units of product A and product B
16 | P a g e

17 | P a g e

Reference
Medhi, J. (2001). Statistical Methods: An Introductory Text (4th ed.). Sydney: New Age
International.
Holsapple, C. & Whinston, B.A. (2013) Decision Support Systems: Theory and Application (6th
ed.). Sydney: Springer Science & Business Media
Lind, A.D., Marchal, G.W. & Wathen, A.S. (2012). Statistical Techniques in Business and
Economics (15th ed.). New York : McGraw-Hill/Irwin.
18 | P a g e