Operations research methodology

Operations Research Methodology is a systematic, scientific approach to decision-making and solving management problems for complex systems. It consists of a sequence of logical stages aimed at identifying a problem, constructing a mathematical or simulation model, finding an optimal or rational solution, and implementing it in practice.

The methodology of operations research (OR) provides a structured analytical process, offering quantitatively-based recommendations to the decision-maker (DM).

Main Stages of the Methodology

While the specific steps may vary depending on the problem and context, the classic operations research methodology typically includes the following main stages:

1. Problem Definition:

• Observation of the system and identification of the problem situation.
• Clear formulation of the research goals and the problem to be solved.
• Definition of the boundaries of the system, its main elements, and interconnections.
• Identification of constraints and resources.
• Definition of the criteria for evaluating the solution's effectiveness.
• This stage often requires close collaboration with stakeholders and subject matter experts.

2. Model Construction:

• Selecting the type of model (mathematical, simulation, graph-based, etc.) that most adequately represents the problem.
• Formalization: translating the problem into the language of mathematics or another formal language.
• Defining variables (controllable and uncontrollable), parameters, and constraints.
• Formulating the objective function, which represents the optimization criterion.
• At this stage, it is crucial to balance the model's realism with its solvability (analytical complexity).

3. Model Solution:

• Applying appropriate operations research methods and algorithms to find a solution within the framework of the constructed model.
• This may involve finding an optimal solution (e.g., using linear programming, dynamic programming) or obtaining characteristics of the system.
• This often requires the use of specialized software.

4. Model Validation and Solution Testing:

• Evaluating the model's adequacy: how well does it represent the real system? Comparing the simulation results with real data (historical or experimental).
• Sensitivity analysis: Investigating how the solution and the value of the objective function change as the model's parameters and assumptions vary. Assessing the solution's robustness.
• If the model or solution is found to be inadequate, the process returns to previous stages (refining the problem, modifying the model).

5. Implementation:

• Interpreting the results of the modeling and the formal solution in a language understandable to the decision-maker (DM) and implementers.
• Developing practical instructions and procedures for implementing the found solution into the real system.
• Training personnel, managing change.
• This stage requires not only technical but also organizational efforts.

6. Control and Follow-up:

• Monitoring the system's operation after the solution has been implemented.
• Evaluating the actual effectiveness of the solution.
• If necessary, adjusting the solution or the model due to changes in external conditions or goals.

Key Principles of the Methodology

The OR methodology is based on the following principles:

• Systems approach: Viewing the problem as part of a larger system, taking interconnections into account.
• Modeling: Using models as the primary tool for analysis and forecasting.
• Optimization: Striving to find the best solution according to criteria and constraints.
• Quantitative basis: Relying on data, measurements, and mathematical methods.
• Interdisciplinarity: Using knowledge and methods from mathematics, statistics, economics, computer science, psychology, and other sciences.
• Decision-making orientation: The ultimate goal of OR is to provide well-founded recommendations for the decision-maker (DM).

Iterative Nature of the Process

The stages of the OR methodology are not always executed in a strictly sequential manner. It is often necessary to return to previous stages to refine the problem, the model, or the data. The operations research process is iterative in nature.

Relationship with Decision-Making

Operations Research does not replace the decision-maker (DM), but rather provides them with scientific tools and quantitatively-based information to make more informed and effective decisions. The optimal solution found using a model is a recommendation that the DM considers alongside other factors (qualitative, strategic, ethical).

Literature

• Taha, Hamdy A. Operations Research: An Introduction. — Pearson. (10th ed., 2017)
• Hillier, Frederick S.; Lieberman, Gerald J. Introduction to Operations Research. — McGraw-Hill Education. (11th ed., 2021)

See Also

• Operations Research
• System Analysis
• Decision Theory
• Modeling
• Mathematical Model
• Optimization
• Objective Function
• Constraints
• Systems Approach
• Sensitivity Analysis
• Model Validation