Decision support Systems

Why DSS???

• Decision Support System is an interactive computer-based information system that, like

MIS, also serves at the management level of an organization which provides information,
models and data manipulation tools to help in making the decision in a semi-structured and
unstructured situation.

• DSS comprises tools and techniques to help in gathering relevant information and analyze
the options and alternatives, the end user is more involved in creating DSS than an MIS.

• it processes information to support the decision making process of managers. It provides

middle managers with the information that enables them to make intelligent decisions.

• Decision support systems are designed for every manager to execute a specific
managerial task or problem. However, sometimes, they can also help in taking complex
decisions.

• it reports past performance and highlights areas that need attention

Example: Financial planning systems, Bank loan management systems.

• A decision support system in a bank, for example, enable a manager to analyze

the changing trends in deposits and loans in order to ascertain the yearly
targets.

• They employ a wide variety of decision models to analyze data or summarize

vast amount of data into a form (usually form of tables or charts) that make
the comparison and analysis of data easier for managers. They provide
interactive environment so that the users could work with them directly, add
or change data as per their requirements, and ask new questions.
Day-to-Day Decision Support System tools and Apllications

1) GPS route planning: determines the fastest and best route between two points by analyzing and comparing multiple
possible options. Many GPS systems also include traffic avoidance capabilities that monitor traffic conditions in real
time, allowing motorists to avoid congestion.

2) Crop-planning tools: farmers use such tools to determine the best time to plant, fertilize and reap as well as weather
forecasting.

3) Medical diagnosis software: that allows medical personnel to diagnose illnesses.

4) Disaster Management tools: DSS provide information about potential hazards, such as floods, cyclones, earth quake
etc. The system includes real-time weather conditions and may include information (current and historical) about
floodplain boundaries and county flood, cyclone and other natural calamities information and prevention techniques.

5) Real estate companies often use DSS for information about properties, including current data such as neighbourhood
comparison prices, property and future planning.

6) Universities and colleges rely on DSS to know how many students are currently enrolled, which allows them to predict
how many additional students are needed in particular courses or overall population to ensure there are enough
students enrolled to meet the university costs.
Decision Support System Examples That Use Historical Data

1) Descriptive analytics: Metrics such as sales results, inventory turnover and revenue growth.

2) predictive analytics: one that helps predict what will happen in the near future. Predictive analytics use a
combination of data mining, statistical tools and machine learning algorithms to determine the likelihood of
certain events taking place. Banks use these techniques to detect fraud, insurance companies use them to
evaluate risk, and ride-hailing firms to determine ticket prices based on demand.(Market Basket Analysis)

3) Diagnostic analytics: Diagnostic information that digs a bit deeper to reveal results and explains reasons for
past performance as measured by descriptive analytics.

4) Business intelligence (BI): Although largely based on historical data, BI solutions allow users to develop and
run queries that are used to guide and support decision-making.

5) ERP dashboards: User-configurable dashboards that allow managers to monitor a variety of performance
indicators.

these tools can predict, based on past performance, external data and market feedback, figures for future product
demand, product obsolescence and returns.
Decision Support System

Conceptual model
Definition: it is a system intended to
support managerial decisions in semi
structured and unstructured decision
situations

DSS were meant to be adjuncts to

decision makers  extending their
capabilities

They are computer based and would

operate interactively online, and
preferably would have graphical output
capabilities
Types of Decisions in an Organization

• Well-defined, clear rules and routine

Structured decisions • Re-ordering inventory, expense reports, payroll calculation

• Some structured elements bust also require judgement and

Semi-structured analysis based on available information
decisions • E.g. Product pricing based on marked analysis and competitor
pricing

• Complex, non-routing and heavily rely on intuition and

Unstructured creative thinking.
decisions • Decision maker’s intuition plays a important role as
information technology offers less support for the decisions

Challenges in semi-structured and unstructured decisions Involve multiple criteria and users have to choose
between conflicting objectives
Types of Decisions in an Organization
Organizational Levels and Types of Decisions

Structured • Well-defined, clear rules and routine

• Re-ordering inventory, expense reports,
decisions payroll calculation

• Some structured elements bust also

Semi- require judgement and analysis based on
structured available information
• E.g. Product pricing based on marked
decisions analysis and competitor pricing

• Complex, non-routing and heavily rely on

intuition and creative thinking.
Unstructured • Decision maker’s intuition plays a
decisions important role as information technology
offers less support for the decisions
Example: A retail company (e.g., Walmart) is experiencing declining
sales in a competitive market

Strategic Decision (Long-term & Unstructured): Should the company

expand to e-commerce?

Tactical Decision (Mid-term & Semi-Structured): Should they introduce

discounts or loyalty programs?

Operational Decision (Short-term & Structured): How should the store

allocate inventory for the next month?
Phases of Decision Making (Simon model 1977)

Intelligence Phase

Design Phase

Choice Phase

Implementation Phase
Four Stages of the Decision Making Process

Stage 1: Intelligence Activity Stage

It is the stage of the decision-making process where some experts identify the problems of an organization.
Based on the issues, the management will try to find a solution that can be applied to the system in order to
provide an effective environment.

Stage 2: Design Activity Stage

This stage involves the study of several strategies which can be applied to find the solution to the problem.
The main characteristic of this stage is the analysis of strategies based on the merits and demerits. The
management analyses the merits and demerits and hence decides which strategy is best suited for the given
problem.

Stage 3: Choice Activity Stage

The third stage gives the decision output. In this stage, the management decides the best-suited strategy,
which can be applied based on the merits and demerits listed in the previous stage. This stage is the
qualitative and quantitative analysis of the strategies and requires creativity and judgemental skills.

Stage 4: Implementation: Implementing the final choice.

Humans consciously or subconsciously follow a systematic decision-
making process - Simon (1977)
1

Simon’s
Decision- 2

Making
Process

4
Hiring process in a company

1.Intelligence: HR collects Design: HR shortlists

applications and job market candidates and plans
trends. interview criteria.

Implementation: The Choice: The best candidate

candidate is onboarded and is selected based on
given training. evaluations.
Decision Making: Intelligence Phase

Problem • Lack of data

Classification • Cost of data collection
• Inaccurate and/or
Potential issues imprecise data
Problem in data / • Data estimation is often
Decomposition information subjective
collection and • Data may be insecure
estimation • Key data may be
qualitative
Problem Ownership
• Data change over time
(time-dependence)
Decision Making: The Design Phase

Finding / developing and analyzing possible courses of actions

A model of the decision-making problem is constructed, tested, and

validated

Modeling: conceptualizing a problem and abstracting it into a quantitative

and/or qualitative form

Developing (Generating) Alternatives

Measuring / ranking the outcomes

 Decision Making: The Choice Phase

The actual decision and the Search Additional

approaches activities
commitment to follow a certain
• Analytic • Sensitivity
course of action are made here techniques analysis
(solving with a • What-if
formula)
The boundary between the design analysis
• Algorithms • Goal seeking
and choice is often unclear (step-by-step
(partially overlapping phases) procedures)
• Heuristics
(rule of thumb)
Includes the search, evaluation, and • Blind search
recommendation of an appropriate (truly random
search)
solution to the model
Decision Making: The Implementation Phase

“Nothing is more
difficult to carry out,
Implementation:
nor more doubtful of Solution to a
putting a
success, nor more problem  Change
recommended
dangerous to handle, management
solution to work
than to initiate a new
order of things.”
Applying Simon’s Decision-Making Model to B-School Admission

Intelligence Phase (Identifying the Problem & Gathering

Information)

Design Phase (Developing Alternatives & Evaluating Options)

Choice Phase (Making the Final Decision)

Implementation Phase (Executing the Decision & Monitoring

Results)
Applying Simon’s Decision-Making Model to B-School Admission

• Understanding the need for an MBA

• Collecting information about B-schools
• Accreditation (AACSB, AMBA, EQUIS)
• Reputation and rankings (Harvard, LBS, IIMs, ISB, etc.)
Intelligence Phase
• Curriculum (General MBA vs. Specialized MBA - Finance,
(Identifying the Problem & Marketing, AI, etc.)
Gathering Information) • Cost & ROI (Tuition fees, scholarships, post-MBA salary)
• Location preferences (Domestic vs. international)
• Exploring eligibility criteria
• CAT/GMAT/GRE scores, Work experience, GPA, Essays, and
Interviews.
Applying Simon’s Decision-Making Model to B-School Admission

• Listing down possible options

Top-tier B-schools (Harvard, Stanford, Wharton, IIMs)
Mid-tier B-schools (Good ROI, but lower global ranking)
Online/Hybrid MBAs (Flexibility but may lack networking
Design Phase opportunities)
(Developing Alternatives & • Weighing the pros and cons of each alternative
Cost vs. ROI
Evaluating Options)
Course flexibility
Placement and alumni network
Industry exposure (Internships, projects)
• Shortlisting 4-5 B-schools based on personal priorities.
Applying Simon’s Decision-Making Model to B-School Admission

Selecting the best B-school based on:

 Personal & career goals (Entrepreneurship, consulting, tech, etc.)

Choice Phase  Admissions probability (Safe vs. ambitious schools)

 Financial feasibility (Scholarships, bank loans)
(Making the Final Decision)  Cultural fit & networking opportunities.
Applying to selected B-schools (finalizing essays, preparing for
interviews.)
Applying Simon’s Decision-Making Model to B-School Admission

Completing the admission process

 Submitting applications
 Attending interviews
 Negotiating scholarships (if applicable)
Implementation Phase Enrolling in the chosen B-school
(Executing the Decision &  Preparing for the MBA journey (Course prep, networking)
Monitoring Results)  Relocating (if necessary)
 Managing finances (education loans, living expenses)
Reviewing and adjusting
 If the first choice doesn’t work, considering backup options (e.g.,
deferring admission, applying next year)
Decision Support Systems (DSS)

Requirements
Examples
• Be interactive and incorporate the
human element as well as hardware DSS for Inventory Management
and software
• E.g. Amazon/Walmart uses DSS for:
• Use internal and external data • Predicting demand based on historical sales data.
• Adjusting stock levels in real-time to avoid
• Include mathematical and statistical shortages or overstock.
models
DSS for Netflix
• Support decision makers at all
• Internal Data: User watch history, ratings.
organizational levels
• External Data: Trending content, seasonal demand.
• Mathematical Models: Algorithms recommend
• Emphasize semi-structured and shows based on user preferences
unstructured tasks
DSS Classifications
Communication-driven and group DSS
• Video Conferencing tools (e.g. Zoom, Webex), Collaborative decision-making systems(e.g.
Think-tank)

Data-driven DSS
• BI tools, Data warehousing, Data mining, OLAP

Document-driven DSS (search engine

• Knowledge management systems (SharePoint, Google Drive), Legal or regulatory
compliance systems

Knowledge-driven DSS
• AI driven recommendation systems, Chatbots, Virtual Assistant

Model-driven DSS
• Optimization tools, Simulation Software. Financial Modelling and forecasting systems
Other DSS Classifications

Institutional and ad-hoc DSS

• ERP decision support modules, Sales forecasting
• Crisis management system, Environment impact analysis system for specific project

Custom-made systems versus ready-made systems

• Custom built supply chain, predictive maintenance
• IBM Watson analytics

Personal, group, and organizational support

• Personal investment portfolio mgmt. tools
• Thinktank
• BI for corporate reporting

Individual support system versus group support system (GSS)

• Salesperson's customer analytics dashboard
• Electronic meeting systems for strategic planning.
Components of DSS

• Includes internal and • Includes mathematical

external data, and a and statistical models
database management that enable a DSS to
system (DBMS) analyze information
• Enables a DSS to
perform data analysis
operations

• Users access the DSS

through user interface
• DSS engine manages and
coordinates the major
components
DSS Capabilities
What-if analysis

Goal-seeking

Sensitivity analysis

Graphical analysis,

Forecasting

Simulation,

Statistical analysis, and

Modeling analysis

29
 A company needs to open warehouses to serve two cities (City A
and City B). They have two possible warehouse locations (W1 and
Problem W2). The objective is to minimize the total cost

Fixed setup
City → A B
cost

W1 ₹5 ₹7 ₹ 10,000
W2 ₹6 ₹4 ₹ 8,000
Demand 100 150

A
Which Total Cost = FC + 𝑣 × 𝑄

warehouse/s
where
FC = Fixed cost
𝑣 = Variable cost per unit
should be W1
𝑄 = Quantity or volume of output

considered to B

serve both cities W2

A company needs to open warehouses to serve two cities (City A and City B).
Total cost
They have two possible warehouse locations (W1 and W2). The objective is A
= 8000 + 100 x 6 + 150 x 4
to minimize the total cost
₹9200

Fixed setup
City → A B B
cost
W1 ₹5 ₹7 ₹ 10,000
W2 ₹6 ₹4 ₹ 8,000
Demand 100 150
W2
Minimize the total cost = Fixed costs of opening warehouses + When warehouse (W2) is considered
Transportation costs.

Total cost
A = 8000 + 10000 + 100 x 5 + 150
x4
A = ₹19100

W1 W1
B
B
When both warehouse are considered
When only W1 is considered W2
Total cost = 10000 + 100 x 5 + 150 x 7 = ₹11550
 A company needs to open warehouses to serve two cities (City A
and City B). They have two possible warehouse locations (W1 and
Problem W2). The objective is to minimize the total cost

Fixed setup
City → A B
cost
W1 ₹5 ₹7 ₹ 10,000
W2 ₹6 ₹4 ₹ 8,000
Demand 100 150
Solution:

A
Only warehouse
(W2) should be
there to serve B
both cities
W2
 Objective function:
M N M
Minimize: ෍ fi . yi + ෍ ෍ tij .xij
i=1 j =1 i=1
Where:
Mathematical 𝑁: Number of cities.
𝑀 : Number of possible warehouse locations.
𝑓𝑖 ​: Annual fixed cost for warehouse i.
Model 𝑦𝑖 : Binary variable (1 if warehouse i is built,0 otherwise).
𝑡𝑖𝑗 ​: Transportation cost from warehouse i to city j
𝑥𝑖𝑗 ​ : Units delivered from warehouse i to city j

Constraints:
Cities demand fulfilment:
σ𝑴
𝒊 𝒙𝒊𝒋 = 𝒅𝒋 , ∀𝒊

Non-negativity and demand limits:

𝟎 ≤ 𝒙𝒊𝒋 ≤ 𝒅𝒋 . 𝒚𝒊 𝒊, j
Binary decision for warehouse construction:
𝒚𝒊 ∈ ሼ𝟎, 𝟏ሽ
Benefits of DSS

Increase in the number of alternatives examined

Fast response to unexpected situations

Ability to make one-of-a-kind decisions

New insights and learning

Improved communication and control over operations

Benefits of DSS

Cost savings from making better decisions and analyze several scenarios in a
short period

Better decisions

Effective teamwork

Time savings

Better use of data resources

Thank You