MIS & Decision Making Concepts

Process and Modeling in Decision Making

There are two basic models in decision making:

• Rational models
• Normative model

The rational models are based on cognitive judgments and help in selecting the most logical and sensible alternative. Examples of such models include – decision matrix analysis, Pugh matrix, SWOT analysis, Pareto analysis and decision trees, selection matrix, etc.

A rational decision making model takes the following steps:

• Identifying the problem
• Identifying the important criteria for the process and the result
• Considering all possible solutions
• Calculating the consequences of all solutions and comparing the probability of satisfying the criteria
• Selecting the best option

The normative model of decision-making considers constraints that may arise in making decisions, such as time, complexity, uncertainty, and inadequacy of resources.

According to this model, decision-making is characterized by:

• Limited information processing: A person can manage only a limited amount of information.
• Judgmental heuristics: A person may use shortcuts to simplify the decision making process.
• Satisfying: A person may choose a solution that is just “good enough”.

Dynamic Decision Making

Dynamic decision-making (DDM) is synergetic decision-making involving interdependent systems, in an environment that changes over time either due to the previous actions of the decision-maker or due to events that are outside of the control of the decision-maker.

These decision-makings are more complex and real-time.

Dynamic decision-making involves observing how people used their experience to control the system’s dynamics and noting down the best decisions taken thereon.

Sensitivity Analysis

Sensitivity analysis is a technique used for distributing the uncertainty in the output of a mathematical model or a system to different sources of uncertainty in its inputs.

From business decision perspective, the sensitivity analysis helps an analyst to identify cost drivers as well as other quantities to make an informed decision. If a particular quantity has no bearing on a decision or prediction, then the conditions relating to quantity could be eliminated, thus simplifying the decision making process.

Sensitivity analysis also helps in some other situations:

• Resource optimization
• Future data collections
• Identifying critical assumptions
• To optimize the tolerance of manufactured parts

Static and Dynamic Models

Static Models:

• Show the value of various attributes in a balanced system.
• Work best in static systems.
• Do not take into consideration the time-based variances.
• Do not work well in real-time systems however, it may work in a dynamic system being in equilibrium
• Involve less data.
• Are easy to analyze.
• Produce faster results.

Dynamic Models:

• Consider the change in data values over time.
• Consider effect of system behavior over time.
• Re-calculate equations as time changes.
• Can be applied only in dynamic systems.

Simulation Techniques

Simulation is a technique that imitates the operation of a real-world process or system over time. Simulation techniques can be used to assist management decision making, where analytical methods are either not available or cannot be applied.

Some of the typical business problem areas where simulation techniques are used are:

• Inventory control
• Queuing problem
• Production planning

Operations Research Techniques

Operational Research (OR) includes a wide range of problem-solving techniques involving various advanced analytical models and methods applied. It helps in efficient and improved decision-making.

It encompasses techniques such as simulation, mathematical optimization, queuing theory, stochastic-process models, econometric methods, data envelopment analysis, neural networks, expert systems, decision analysis, and the analytic hierarchy process.

OR techniques describe a system by constructing its mathematical models.

Heuristic Programming

Heuristic programming refers to a branch of artificial intelligence. It consists of programs that are self-learning in nature.

However, these programs are not optimal in nature, as they are experience-based techniques for problem solving.

Most basic heuristic programs would be based on pure ‘trial-error’ methods.

Heuristics take a ‘guess’ approach to problem solving, yielding a ‘good enough’ answer, rather than finding a ‘best possible’ solution.

Group Decision Making

In group decision-making, various individuals in a group take part in collaborative decision making.

Group Decision Support System (GDSS) is a decision support system that provides support in decision making by a group of people. It facilitates the free flow and exchange of ideas and information among the group members. Decisions are made with a higher degree of consensus and agreement resulting in a dramatically higher likelihood of implementation.

Following are the available types of computer based GDSSs:

• Decision Network: This type helps the participants to communicate with each other through a network or through a central database. Application software may use commonly shared models to provide support.
• Decision Room: Participants are located at one place, i.e. the decision room. The purpose of this is to enhance participant’s interactions and decision-making within a fixed period of time using a facilitator.
• Teleconferencing: Groups are composed of members or sub groups that are geographically dispersed; teleconferencing provides interactive connection between two or more decision rooms. This interaction will involve transmission of computerized and audio visual information.