What Are the Different Types of Decision Analysis Software?

Decision analysis generally refers to a quantitative analysis method that chooses one of several decision-making processes through decision analysis techniques, such as expectation method or decision tree method. It is mainly used in dynamic meteorology and other disciplines in atmospheric science.

Decision analysis generally refers to a quantitative analysis method that chooses one of several decision-making processes through decision analysis techniques, such as expectation method or decision tree method. It is mainly used in dynamic meteorology and other disciplines in atmospheric science.

Chinese name

Strategic analysis

Foreign name

decision analysis

Explanation

Decision Analysis Technology

Apply to

Dynamic meteorology and other disciplines in atmospheric science

Use mathematical planning, etc.

Decision subject, decision objective, etc.

Introduction to Decision Analysis

Most decision-making theories are normative, that is, decision-theory theory achieves optimal decision-making by assuming an ideal decision-maker with complete information, accurate calculations, and complete rationality (in practice, some so-called " The "best" scenario is not the maximum, and the optimal may also be contained in a specific or approximate maximum). The practical application of this canonical model (how people should make decisions) is called decision analysis, and its goal is to help people make tools and methodologies for further good decisions. Decision support system is a systematic and comprehensive software tool developed in this way. Because people usually do not agree with axioms, they often violate their optimality. Related research on this phenomenon is called descriptive discipline. This descriptive model attempts to describe how people actually do it. Because normative and optimal decisions often test hypotheses against people's actual actions, normative and descriptive models are associated. Allowing further testing of decisions that occur in practice may loosen constraints on complete information, rationality, and other methods in the canonical model. In recent decades, more and more researchers have sparked interest in what is known as "behavioral decision theory," which has contributed to re-evaluating the requirements of rational decision theory. ^[1]

Decision analysis steps

Decision analysis generally consists of four steps: (1) forming a decision problem, including proposing a plan and determining a goal; (2) judging the natural state and its probability; (3) drawing up multiple feasible plans; (4) evaluating the plan and making a choice . Commonly used decision analysis techniques are: decision analysis in a certain situation, decision analysis in a risk situation, and decision analysis in an uncertain situation. ^[1]

(1) Decision analysis in a deterministic situation. The main characteristics of a deterministic decision-making problem are four aspects: first, there is only one state, second, there is a clear goal that the decision maker wants to achieve, third, there are two or more options that the decision maker can choose, It is clear that the return value of different schemes in this state. Deterministic decision analysis techniques include the use of differential methods to find maximum values and mathematical programming. ^[1]

(2) Decision analysis in risky situations. This type of decision-making problem differs from deterministic decision-making only in the first point: in a risk-based situation, there may be more than one possible state in the future. Which state will appear in the future cannot be confirmed in advance, but only the possibility of various states. Size (such as probability, frequency, scale, or weight, etc.). The commonly used risk-based decision analysis techniques include expectation method and decision tree method. The expected value method is based on the probability average value of each feasible scheme in each natural state, and determines the choice of each scheme. The decision tree method is helpful for decision makers to visualize decision problems. It can simply draw a variety of replaceable schemes, possible states, possibilities, and consequences, etc. on a map for calculation, research and Analysis, can also be added and corrected at any time. ^[1]

(3) Decision analysis under uncertainty. If there is more than one state, and the probability of the occurrence of each state is unknown, it is called an uncertain decision. Common decision analysis methods are: a. Optimism. More optimistic decision makers are willing to fight for every opportunity to get the best results. The decision step is to select a maximum return value from each plan, and then select a maximum value from these maximum return values, and the plan corresponding to the maximum value is the selected plan. b. Pessimism. More pessimistic decision makers are always cautious and think of the worst outcome. The decision step is to first select a minimum return value from each plan, and then select a maximum return value from these minimum revenue values, and the corresponding plan is the optimal plan. This is to find the most favorable solution from all kinds of the most unfavorable situations. c. Equal likelihood criteria. Decision makers have no knowledge of state information, so they treat them equally as if they were equally likely. This allows decisions to be made in a risk-like manner. ^[1]

Decision Analysis Features

Decision analysis in deterministic situations. The main characteristics of a deterministic decision-making problem are four aspects: one is that there is only one state, and the other is that there is a clear goal that the decision-maker wants to achieve. The third is that there are two or more schemes for decision makers to choose from, and the fourth is that the return value of different schemes in this state is clear. Deterministic decision analysis techniques include the use of differential methods to find maximum values and mathematical programming. ^[1]

Constituent elements

Decision analysis is a comprehensive subject that is closely related to economics, mathematics, psychology, and organizational behavior. Its research object is decision-making, and its research purpose is to help people improve the quality of decision-making and reduce the time and cost of decision-making. Therefore, decision analysis is a creative management technique. Decision analysis includes the processes of finding problems, determining goals, determining evaluation criteria, formulating plans, selecting plans, and implementing plans. Decision analysis usually has the following components: ^[1]

(1) Decision-making subject

Decisions are made by people, and people are the subject of decisions. In the process of decision analysis, the decision-making subject who only undertakes tasks such as asking questions or analyzing and evaluating solutions is called "analyst", and in the process of decision analysis, the decision-making subject who can make the final decision is called "leader". ^[1]

(2) Decision objectives

Decisions must have at least one desired goal. Decisions revolve around goals. The beginning of a decision is to determine the goal, and the terminal is to achieve the goal. Decision-making goals reflect both the subjective will of the decision-making subject and objective facts. Without goals, decision-making is impossible. ^[1]

(3) Decision plan

Decisions must have at least two options available. There are two types of schemes:

Clear plan has a limited number of clear specific plans;

The ambiguous plan only explains the possible constraints for generating the plan. The number of plans may be limited or unlimited. ^[1]

(4) The ending

The ending is also called the state of nature. One or several possible outcomes may occur after the implementation of each scheme. If there is only one outcome in each scheme, it is called a "deterministic" decision; if each scheme produces at least two possible outcomes, it is called " "Risk" or "Uncertain" decisions. ^[1]

(5) Utility

The value of each outcome of each scenario is called utility. Evaluate the pros and cons of the scheme according to the utility value of each scheme. ^[1]