Since the 1950s, engineering design techniques have referenced means-ends analysis (MEA). The means-ends chain method, common in consumer behaviour analysis, is also connected to MEA. Beginning a mathematical proof helps to organize one's thoughts this way.

Objective

The means-ends analysis limits fruitless research when finding a solution to a problem. The gap between the existing and desired states is minimized by selecting an action that brings about the latter. A measure of the present state creates a new state, repeated recursively to reach the desired state. Keep in mind that the goal-seeking system needs to link the behaviours pertinent to mitigating any form of observable difference if MEA is to be effective. 

To further boost the average performance of MEA over other brute-force search algorithms, an essential difference is chosen first when knowledge about the relevance of differences is available. MEA outperforms other search heuristics in the average situation by focusing on the fundamental differences between the present state and the objective rather than on relevance.

How does it work?

Recursively applying the means-ends analysis approach to an issue is possible. It is a tactic for managing search when addressing problems. The primary steps that describe how the MEA technique works to solve an issue are as follows.

• Analyze the differences between the initial state and the end state first.
• Choose from a range of operators that it can use to account for each difference.
• The difference between the current and intended states is decreased by using the operator at each difference.

Conclusion

Allen Newell and Herbert A. Simon first introduced the MEA technique as a problem-solving technique in their 1961 computer programme General Problem Solver (GPS). The system is informed of the relationship between differences and operators in such an approach. (In GPS, this information was presented as a table of connections.) Moreover, when an operator's activity and unintended consequences are obvious, the search can choose the pertinent operators without using a table of links by looking at themselves. This latter scenario permits task-independent correlation of differences to the operators, which reduces them, with the automated planning computer software STRIPS serving as the typical example.

Jaime Carbonell, Steven Minton, and Craig Knoblock's learning-assisted automated planning initiative at Carnegie Mellon University produced Prodigy, a problem solver utilized by MEA. Furthermore, the Technical University of Denmark's Professor Morten Lind created a programme called Multilevel Flow Modeling (MFM). It conducts diagnostic reasoning based on means-ends for industrial control and automation systems.