What Is the Analytic Network Process?

Diplomatic and military analysts, civic planners, and corporation executives are often faced with making decisions that have numerous, somewhat opaque, parameters. They can use a multicriteria decision analysis tool known as the Analytic Network Process (ANP) to assist. ANP gives more than cost-benefit analyses; it makes use of four elemental criteria: benefits (B), opportunities (O), costs (C), and risks(R), from which decision-makers can model the sub-elements of a decision. These four elements are examined against weighted priorities and produce rankings of alternatives in a synthesized framework of what-if-scenarios. These scenarios can then be viewed against objectives to determine the alternative that will facilitate the best outcomes.

The analytic network process initially makes use of organizing hierarchical networks of the objectives and the elements of these objectives. At the same time, evaluations of the influences, feedback connections, and interdependencies of each of the four elements and their sub-elements are mapped. In these evaluations, benefits and opportunities can receive valuable balance in considering the costs and risks using a (B+O) minus (C+R) approach. These results are further fleshed out by use of ratio scale measurements while making pair wise comparisons between the elements.

Ratio scale measurements are developed that assign weight to priorities made up of not only objective facts, but also subjective data. Subjective data can be factors such as nodes of societal beliefs and values, political biases and agendas, and both tangible and intangible criteria, each weighted according to their priority positions. The synthesis of these objectives’ criteria and subjective positions are what form the basis of the rankings of alternatives in what is called a super-matrix.

In application, the balancing of all these elements gives a big picture presentation of each what-if-scenario and of the alternatives’ rankings in the super-matrix. Predicting these outcomes using analytic network process entails both quantitative and qualitative analysis and deductive and inductive reasoning to forecast the probable determinants and likely results. The what-if-scenarios are subjected in their final analysis to sensitivity analysis, so that any changes in judgment, design, or priority inputs can test their stability.

Many real world examples of application of analytic network process have been reported in several books and published reports. For example, a Turkish construction corporation reported its use of analytic network analysis and its BOCR criteria scenarios when considering bidding on an overseas construction project. The company needed to consider risks such as an elevated level of terrorism and different cultural, political and legal environments surrounding the project in the other country. Costs such as unavailability of certain raw materials that would need to be imported further weighed against any benefits or opportunities realized from the project. These were only a few of the many factors that ANP modeled to determine if the benefits and opportunities of the project would outweigh the costs and risks.