Systems Theory/Decision Structure

Decisions
What is a decision? Merriam- Webster’s dictionary defines a word as a noun, with meaning “a position arrived at after consideration.”

Mallach, defines a decision as a reasoned choice among alternatives. Examples of this are:
 * Where to advertise a new product,
 * What stock to buy,
 * What movie to see,
 * Where to go for dinner.

Some other authors define the term in similar manner, but with more obvious mathematical approach to definition (Zvi Covaliu, 2001, A Fair Isaak White Paper). For Covaliu, a decision, refers to a point in time when the decision maker has to choose one alternative, out of a domain of available alternatives, that could be discrete or continuous, simple or combined. An alternative can be a combination of values to be chosen at the decision point for a number of decision variables.

Different Decisions
What separates one decision from another is the difference in the information set available to the decision maker before each decision is made. The information set corresponding to a decision is the set of all observations available to the decision maker prior to making that decision.

Hammond, Kenney and Raiffa equate the decision with, what they call, “smart choice.” According to them, every “smart choice” can be broken down to eight elements:


 * Problem
 * Objectives
 * Alternatives
 * Consequences
 * Trade-Offs
 * Uncertainty
 * Risk Tolerance
 * Linked Decisions

First five of these elements are applicable to virtually any decision. Last three- uncertainty, risk tolerance and linked decisions- help clarify decisions in volatile or evolving environments. Some decisions won’t involve these elements, but many most important decisions will.

Mallach, on the other hand defines decision elements in different terms. For him, the decision comprises of three elements:


 * Decision Statement- What are we trying to decide?
 * Alternative- What are the options?
 * Decision Criteria- How are we going to judge the merits of each alternative?

Further, the decision can be classified in three types, based on the type of structure or nature of the task:


 * Structured Problems
 * Routine and repetitive with standard solution
 * Well defined decision making procedure
 * Given a well-defined set of input, a well defined set of output is defined
 * Semi-structured Problems
 * Has some structured aspect
 * Some of the inputs or outputs or procedures are not well defined
 * Unstructured Problems
 * All phases of decision making process are unstructured

Simon, on the other hand, classifies decisions into two categories:
 * Programmed decisions, and
 * Non-programmed decisions.

The programmed decision is repetitive and routine. The non-programmed decisions are novel, unstructured, and call for intelligent adaptive and problem oriented action when solving the problem. Simon states that the traditional techniques for these non-programmed decisions are judgment, creativity, and intuition.

C.I. Barnard categorizes decisions into logical and non-logical. Logical decision making involves conscious thinking and reasoning, and the process is expressible in words or other symbols. Non-logical decision making is not capable of being expressed in words, or as reasoning, and is made known only by the action itself.

The decisions are made in order to solve a particular problem. According to Duncker (1945) solving a problem means continuous narrowing of the range of options through modifications of representations. This repetitive process may continue until there may be only one alternative left to consider. Hence, it is not so strange to see “decision-making” and “problem solving” used interchangeably in common language and discussion. As stated in the paragraph above, decision has to be reached; it has to be made through a deliberate process of decision-making. As such, once the term is used, certain action is expected on behalf of the agent that makes a decision (decision-maker). According to Simon, the decision-making process consists of three main stages:


 * 1) Intelligence: Fact finding, problem and opportunity sensing, analysis, and exploration.
 * 2) Design: Formulation of solutions, generation of alternatives, modeling and simulation.
 * 3) Choice: Goal maximization, alternative selection, decision making, and implementation.

Other authors also consider the implementation to be integral element of the decision-making process, building upon Simon’s observations.

For Hammond et al, in order to make the right decision, one must:


 * 1) work on the right decision problem
 * 2) specify the objectives
 * 3) create imaginative alternatives
 * 4) understand the consequences
 * 5) grapple with trade-offs
 * 6) clarify the uncertainties
 * 7) think hard about risk tolerance
 * 8) consider linked decisions

Decision making can be divided into 3 types: strategic, management control and operations control. Within each of these levels, decision making can be classified as either structured or unstructured, closely paralleling the Simon’s classification of the decision (i.e. structured decision making uses programmed decisions).

Decisions are difficult. There are four key areas that determine the relative difficulty of a decision:
 * Structure – in general, the more structure, the less information required
 * Cognitive limitations – the human mind is limited to handling 5 to 9 distinct pieces of information
 * Uncertainty – the amount is based on how complete and accuracy of the information
 * Alternatives and multiple objectives – the selection of one alternative may impede the progress towards a different goal

There are two traditional models of decision making: Simon famously discussed limits on decision making. According to him, there are two limits: Influenced by these limitations, decision makers are often forced to make decisions that are just good enough. This is what Simon referred to as “satisficing.”
 * 1) The rational man model is associated with economics. The rational decision-maker systematically searches for the decision that will maximize results. Goal is to optimize or maximize.
 * Know the goals, know the goal that optimizes.
 * Know all the alternative courses of action.
 * Know the relation of courses of action to goals, this is you can predict consequences of actions, i.e. can predict outcomes of actions.
 * 1) Alternative model called the administrative model associated with Herbert Simon. Simon says the rational man model is prescriptive or normative, the way it is supposed to be, rather than the way it is.
 * 1) Bounded rationality: decision-maker receives imperfect information about goals and courses of action and relation of means to ends. Decision-making is then based on these restrictions.
 * 2) Bounded discretion: constraints on optimizing, prior commitments, moral and ethical standards, laws, and social standards.