DSS Chapter 2 – Foundations and technologies for decision making

Groupthink can lead to bad decisions.

- Decision makers are interested in evaluating what-if scenarios.

• Experimentation: (challenges)

- The real system may result in failure and time time-consuming.

- With the real system is possible only for one set of conditions at a time, and can be
disastrous.

• Changes (challenges)

In the decision-making :

1. environment may occur continuously, leading to invalid assumptions about the situation.
2. Environment may affect decision quality by imposing time pressure on the decision
maker.
- Collecting info and analyzing a problem takes time and can be expensive. (difficult to
determine when to stop and make a decision).

DSS 3 Main concepts:

1. Support 2. Decision making 3. System
- Building successful dss requires a good understanding of these 3 concepts.
- Support through : a rational approach that simplifies reality and provides a relatively
quick and inexpensive means of considering alternatives.

Decision making: a process of choosing among 2+ alternative for the purpose of

attaining a goal(s).

- Managerial decision making is synonymous with the entire mgt process. (simon, 1977)

Stages in Decision Making (simons 4 steps of the decision making)

(a process, judgement, reasoning):

1. Intelligence ( find what to fix)

2. Design (find fixes)
3. Choice (pick a fix)
4. Implementation (apply the fix)

Problem solving (a method, analytical aspect of thinking)

Decision-Making Disciplines:

1. Behavioral : Anthropology, law , philosophy, political science, psychology , sociology.

2. Scientific: computer science, decision analysis , economic , engineering , hard sciences

- Each discipline:
1. Has its own set of assumptions
2. Contributes a unique, valid view of how ppl make decisions.

Quality factors of decision and decision-making:

- Accuracy (effectiveness) - comprehensiveness -fairness
- Speed (efficiency) -coherence -due process (legal procedures)

Decision style (how decision makers think and react to problems):

How they:

- Perceive a problem (awareness)

- Cognitive response (thinking)
- Values and beliefs (ethics)

Decision making styles:

1. Heuristic (experimental/investigative) VS analytic

2. Autocratic VS democratic
3. Consultative (with individuals or groups)

Phases of decision making process (simons decision making process)

(systematic decision
making process)
1. Intelligence
2. Design
3. Choice
4. Implementation
5. Monitoring

-simons model is the most

concise and yet complete
characterization of rational
decision making.
 1. Intelligence Phase
- Scan the environment, either intermittently or continousyl.
- Identify problem situations or opportunities
- Monitor the results of implementation
- Problem is the diff btwn what ppl desire (or expect) and what is actually occurring.
Problem= Desired – Actual > 0

(timely identification of opportunities is as imp as identification of problems)

• problem classification

- Classification of problems according to the degree of Structuredness

• problem decomposition

- Often solving the simpler subproblems may help in solving a complex problem
- Info/data can improve the structuredness of a problem situation

• problem ownership

- Assign responsibility

- Outcome of intelligence phase = A formal problem statement

2. Design Phase

(finding/developing and analyzing possible courses of actions (solutions))

- 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 (using symbols/ variables) (both an art and a science)
Abstraction : making assumptions for simplification

The design phase: Decision Variables

1. Input variables: the alternatives among which a manager must choose.

2. Result Variable(s): that describes the objective or goal of the decision making problem
3. Uncontrollable variables or parameters: describe the environment
- The process of modeling involves determining the relationships among the variables;
usually mathematical, sometimes symbolic.
 Selection of a principle of choice:

It is a criterion that describes the acceptability of a solution approach.

- Reflects decision-making objective(s)

- In a model, it is the result variable reaching the acceptable criterion

Choosing & validating (building the right model) against:

- High risk vs low risk

- Optimize vs sacrifice
(criterion is not a constraint)

The design phase : 3. types of principles of choice

1. Normative models (optimization)

- The chosen alternative is demonstrably the best of all possible alternatives. This is
typically the goal of prescriptive analytics.
Assumptions of rational decision makers:
- Objective is to max. the attainment of goals
- All alternatives courses of action and consquences are known
- Preference to rank the desirability of all consequences

2. Heuristic models (suboptimization)

- The chosen alternative is the best of only a subset of possible alternatives; w/o
considering the details of the rest of the org, then an optimal solution from the point of
view of that part may be inferior for the whole.
- Rational : often it is not feasiable to optimize realistic problems (due to size/complexity)
- Suboptimization may also help relax unrealistic assumptions in models.
- Help reach a good enough solution faster.

3. Descriptive models
- Describe things as they are or as they are believed to be
- They don’t provide a solution but info that ay lead to a solution
- Simulation : most common descriptive modeling method (allows exoirementation with
the descriotive model)
- Narrative model : a story that helps a decision maker uncover the imp aspects of the
situation and leads to better understanding and framing
 4. Good enough or Satisficing (smth less than the best)
- A form of suboptiminzation
- Does not aim to evaliuate all or even a subset of the best alternatives but rather stops once
a solution that meets the minimum acceptable criteria is found.
- Seeking to achieve a desired level of performance as opposed to the best which is known
as trade off
- benefit is time saving decisions may lose value over time
- marginal benefit of a better solution versus the marginal cost

differences between Heuristic Models (suboptimization and good enough or satisficing)

1. Heuristic models sub-optimization focus on finding the best solution within a restricted subset
often ignoring the entire systems optimality while satisficing focuses on achieving a solution that
meets acceptable criteria quickly without seeking the best solution.

2. Heuristics aim to simplify by limiting the scope of a search whereas satisficing aims to stop
the search sounds good enough solution is found

3. Heuristics may still aim for a locally optimal solution, within the subset satisficing does not
pursue optimality at all and prioritizes efficiency over perfection

The design phase : Generating alternatives

In optimization models example linear programming the alternatives may be generated

automatically.

In most DSS situations however it is necessary to generate alternatives manually but lengthy
processes costly need expertise in the problem area.
Use of GSS helps generate outcomes.
Measuring/ranking the outcomes.

- Using the principle of choice can be multi criteria

- Sometimes an outcome is expressed directly in terms of a goal

Uncertainity: lack of precise knowledge

Scenario (what-if case) : a statement of assumptions about the operating environment of a

particular system at a given time
Decision making : the design phase

Decision Making: The choice phase

- The actual decision on the commitment to follow a certain course of action are made
here.
- The boundary between the design and choice is often unclear partially overlapping phases
due to:
1. Generate alternatives while performing evaluations
2. Includes the search for , evaluation of and recommendation of an appropriate solution
to the model.
3. Choices can be evaluated as to their viability and profitability.

Search approaches for an appropriate course of action include:

1) Analytic techniques (solving a formula).

2) Algorithms (step-by-step procedures).

3) Heuristics (Investigative/ Experimental).

4) Blind search (truly random search).

▪ Additional activities, to determine the robustness (strength) of any given alternative:

1) What-if analysis, major changes.
 2) Sensitivity analysis, slight changes in the parameters.

3) Goal seeking, helps to determine values of the decision variables to meet a specific
objective

4. The implementation phase

Solution to a problem = change

Implementation: putting a recommended solution to work

- The decision-making process, though conducted by people can be improved with

computer support.

How are decisions supported ? (support for the design phase)

- To generate alternative courses of action, determining the critera for choice.

Generating alternatives:

1. Structured/simple problems: use standard and/or special models

2. Unstructured/complex problems: use human experts, kms, brainstorming, olap,
generative ai and advanced machine learning.

Dss capabilities

DSS early definition: it is a system intended to support managerial decisions in

semistructured and unstructured decision situations.
DSS were meant to be aides to decision makers: → extending their capabilities but not
replacing their judgment.
They are computer based and would operate interactively online and preferably would have
graphical output capabilities.

▪ Nowadays, simplified via Web browsers and mobile devices.

DSS classifications

(AIS SIGDSS1 has adopted a concise classification scheme for DSS that was proposed by
Power (2002))

1. Communication-driven and Group DSS

2. Data-driven; OLAP and reporting analytics s/w systems.

3. Document-driven; e.g., Most KMS.

4. Knowledge-driven; e.g., ANN and ES.

5. Model-driven; e.g., Microsoft Excel, financial models (FinTech).

- often DSS is a hybrid of many of the above.

Components of DSS

1. Data Management Subsystem.

- Includes the database that contains the data.

-Managed by s/w: Database Management System (DBMS).

-Can be connected to the corporate DW.

2. Model Management Subsystem.

- Model Base Management System (MBMS).

3. Knowledgebase Management Subsystem.

-Organizational knowledge base (e.g., MyGJU).

4. User Interface Subsystem.