Network Analysis Process (Analytic Network Process, abbreviated as ANP) is a multi-criteria decision analysis method that helps decision-makers effectively evaluate and rank various options in complex decision environments.
Compared to traditional Analytic Hierarchy Process (AHP) (Previous article: What is AHP? Is it suitable for your research?), ANP can handle not only hierarchical structures but also network structures, making it highly valuable in multi-level and interdependent decision problems.
Network Analysis Process (ANP)
The Network Analysis Process (ANP) was proposed by American scholar Thomas L. Saaty. ANP extends the principles of the Analytic Hierarchy Process (AHP) and is suitable for handling situations where there are complex interdependencies and feedback loops among decision factors. ANP reflects these dynamic relationships among decision elements by constructing a network model, thus providing a more comprehensive and flexible analytical approach.
The core principle of ANP is to transform the decision problem into a directed network composed of nodes and edges. Nodes represent various elements of the decision problem (such as criteria, sub-criteria, options, etc.), while edges represent the dependencies between these elements. By constructing such a network, ANP can quantify the relative importance of each element and calculate optimal decision options.
Therefore, ANP not only aligns more closely with human thought patterns but also transforms the originally standardized hierarchical structure into a complex network akin to “amoebas”, allowing researchers to more effectively describe the characteristics of the problem, even enabling them to ponder logical issues like “which came first, the chicken or the egg?”.
Steps of the Network Analysis Process (ANP)
The calculation process of ANP mainly includes the following steps:
Step 1: Construct the network model
When conducting network analysis, you first need to clarify the problem and set the goals to be achieved. Based on the goals, evaluate the criteria group dimensions and sub-criteria, and further establish the interdependencies, internal correlations, and overall mutual influences among the criteria to construct a directed network.
Step 2: Define the assessment scale
Saaty (1971) proposed measuring the weights of criteria on a scale of 1 to 9, allowing the semantic values of relationship influences to be defined according to semantic operational types.
Step 3: Construct pairwise comparison matrices, calculate eigenvalues and vector values
Using the network relationships formed by the dependencies or self-feedback effects among the criteria, establish pairwise comparison matrices based on the results evaluated in Step 2, and calculate the eigenvalues and vector values.
Step 4: Consistency check
Considering that there may be differences in expression or inconsistencies in results when answering questions, it is necessary to check the consistency of the decision-maker’s judgments using the Consistency Ratio (CR) to avoid such issues.
Step 5: Calculate the unweighted supermatrix
Through the pairwise comparison matrix that passed the consistency check, a large matrix is obtained after calculating the eigenvector values, which can be used to calculate the influence weights among the criteria and integrate the eigenvectors of each dimension into an unweighted supermatrix Y (as shown in the figure below). This unweighted supermatrix indicates the range of influence of internal factors of a dimension on other dimension factors, and if it has no influence on itself or on other factors, it is represented by 0.
Step 6: Establish the cluster dimension priority matrix
Before calculating the weighted supermatrix, the expert responses regarding the priority values among the cluster dimensions need to be organized and a cluster dimension priority matrix FD established.
Step 7: Establish the weighted supermatrix
Multiply the cluster dimension priority matrix FD by the unweighted supermatrix (Y) to obtain the weighted supermatrix, with the sum of each row in the weighted supermatrix equal to 1.
Step 8: Calculate the limit supermatrix
The weighted supermatrix is taken to the limit, meaning that the weighted supermatrix Ya is self-multiplied until it reaches a convergent and stable extremum, which will remain fixed and unchanged; this matrix is called the limit supermatrix.
Applications of ANP
The Network Analysis Process (ANP) has wide applications across various fields and industries, especially in the following areas:
Project Management: Used for project selection, prioritization, and resource allocation.
Supply Chain Management: Used for supplier evaluation, logistics optimization, and risk management.
Product Design: Used for product feature analysis, market positioning, and competitive strategies.
Marketing: Used for target market selection, marketing strategy formulation, and brand positioning.
Strategic Planning: Used for goal setting, resource allocation, and execution planning.
Educational Models: Theory of Planned Behavior (TPB) | Technology Acceptance Model (TAM) | Unified Theory of Acceptance and Use of Technology (UTAUT) | Expectation Confirmation Theory (ECT) | Diffusion of Innovations Theory (DI) | Theory of Reasoned Action (TRA) | Hedonic Motivation System Acceptance Model (HMSAM) | Impression Management Theory (TIM) | Service Quality | Task-Technology Fit (TTF) | Protection Motivation Theory (PMT) | Process Virtualization Theory (PVT) | Information Systems Success (D&M-IS success)
Sociological Models: Theory of Planned Behavior (TPB) | Technology Acceptance Model (TAM) | Unified Theory of Acceptance and Use of Technology (UTAUT) | Expectation Confirmation Theory (ECT) | Diffusion of Innovations Theory (DI) | Theory of Reasoned Action (TRA) | Hedonic Motivation System Acceptance Model (HMSAM) | Impression Management Theory (TIM) | Service Quality | Task-Technology Fit (TTF) | Process Virtualization Theory (PVT) | Information Systems Success (D&M-IS success)
Management Models: Theory of Planned Behavior (TPB) | Technology Acceptance Model (TAM) | Unified Theory of Acceptance and Use of Technology (UTAUT) | Expectation Confirmation Theory (ECT) | Diffusion of Innovations Theory (DI) | Theory of Reasoned Action (TRA) | Hedonic Motivation System Acceptance Model (HMSAM) | Impression Management Theory (TIM) | Service Quality | Task-Technology Fit (TTF) | Protection Motivation Theory (PMT) | Multi-Motivation Information Systems Continuance Model (MISC) | Process Virtualization Theory (PVT) | Information Systems Success (D&M-IS success)
Recommended Scales: Social Anxiety | Health Survey | New Environment Concern (NEP) | Activities, Interests, Opinions (AIO) | Overall Well-being | Technology Readiness | Depression, Anxiety, and Stress | State Anxiety | Depression | Social Avoidance and Distress | Self-efficacy | Leadership Behavior Description | Multifactor Leadership | Transformational Leadership | Transactional Leadership | Psychological Well-being | Subjective Well-being | Service Quality | Learning Motivation | Learning Styles | Organizational Citizenship Behavior | Achievement Goals | Work Engagement | Social Capital | Impression Management | Psychological Skills | School Belonging | Artistic Self-perception | Travel Experience | Psychological Empowerment | Life Satisfaction | Brand Image | Course Experience | Turnover Intention | Input-Output (Feedback) | Silence Behavior | Job Insecurity | Organizational Culture | Organizational Goal Ambiguity | Implicit Followership | Organizational Justice | Self-control | Intolerance of Uncertainty | Anxiety Self-assessment | Depression Self-assessment | Pennsylvania Worry | Post-Traumatic Growth | Organizational Culture Identity | Perceived School Atmosphere | Psychological Quality | Work Prosperity | General Procrastination | Event Impact | Parental Low-head Behavior | Idol Worship | Teaching Quality Evaluation | Positive Affect/Negative Affect Scale | Psychological Resilience Scale | Simplified Self-control Scale | Impulsivity Scale | Self-compassion Scale | Self-compassion Scale Short Version | Overall Brand Equity Scale | Brand Identity Scale | Brand Attachment Scale | Brand Value Scale | Employee Voice Behavior Scale | Self-perceived Burden Scale | Innovation Performance Scale | Creative Self-efficacy Scale | Learning Adaptation Scale