Quantitative analysis is the scientific approach to managerial decision making whereby raw data are processed and manipulated resulting in meaningful information. It offers rational mathematical approach that may generate several alternatives while avoiding whim or judgmental criteria.
There are several steps in the quantitative analysis process.
1. Define the problem – This involves formulating a clear and concise statement that gives direction and meaning to the subsequent QA steps and requires specific, measurable objectives. This is often the most difficult and most important step.
2. Develop a model – A model is a representation (usually mathematical) of a situation. In terms of quantitative analysis, this model will provide a realistic, solvable, and understandable mathematical statement showing the relationship between variables. Models contain both controllable (decision variables) and uncontrollable variables and parameters. Typically, parameters are known quantities (salary of sales force) while variables are unknown (sales quantity).
3. Acquire input data – once a model is developed, data must be obtained to use in the model – input data. Obtaining accurate input is vital since improper date may result in misleading results – aka, garbage in / garbage out. Useful data may come from a variety of sources such as company reports, company documents, interviews, on-site direct measurement, or statistical sampling.
4. Develop a solution – Manipulating the model variables until a practical and implementable solution is obtained is how the optimal model solution is found. Manipulation can be done by solving the equation(s), trying various approaches (trial and error), trying all possible variables (complete enumeration), and/or implementing an algorithm (repeating a series of steps). The accuracy of the solution depends on the accuracy of the input data previously described.
5. Test the solution – Thorough testing is necessary prior to analyzing and implementing a solution. One means of testing the model and the input data involves collecting data from a different.. An example would be if the original data was gathered in a survey, additional data may be gathered through direct measurement or sampling.
6. Analyze the results and Sensitivity Analysis – Since most solutions will result in some kind of action or change in the way a firm operates, understanding their implications, as well as conducting a sensitivity analysis (a change to input values or the model) to evaluate the impact of a change in model parameters. Sensitivity or post-optimality analyses determine how the solution(s) will change with a different model or input data.
7. Implementing the results – The final step in the QA process is implementation of the solution into the company and the monitoring of the results as changes may evolve which require a reworking of the original model.