Introduction to Quantitative Analysis

Problem #1:

As discussed in Week 1 lecture, Introduction to Quantitative Analysis, qualitative examples of problems faced by engineering managers can range from state and federal legislation, technological breakthroughs, social and political incentives, and personal relations. Arguably, two of these four categories can be digested and interpreted in another form of analysis.

For example, can a qualitative factor such as new legislation be represented in a true/false definition? If an outcome such as this can yield two different scenarios, both could be mapped and planned for equally. It could then be said that this factor, although qualitative, can be represented quantitatively as a percentage of likelihood (in this case 50%) for managers planning business strategy based on the outcome, effectively being outlined as a future probabilistic model.  

A similar approach could be applied to social and political incentives, if the incentives are in the form of monetary deductions or reimbursement. Consider the case of Tesla and the $7,500 tax rebate their customers receive from the US government upon the purchase of a new fully-electric vehicle. The company applies this incentive quantitatively against market prices of competitor’s vehicles to forecast yearly sales. The legislation is also written with discrete and explicit allowance terms. Once a company, such as Tesla, sells 200,000 qualifying electric vehicles in the US, the tax credit will start to expire. Then the credit is cut in half in the second calendar quarter after the company sells the 200,000th vehicle. Six months after that, the credit is reduced to 25% of its original value. Six months after that, the credit is eliminated completely (Matousek, 2018). This political incentive is described quantitatively which gives managers the ability to apply accurate data to form deterministic models to compare number of sales versus incentive for prospective buyers.

There are many ways to mold qualitative factors into quantitative data which makes decision making easier for companies, which is a good thing. However, the question is, where might we see purely qualitative problems. A perfect example is one of the four categories mentioned above, personal relations. Let’s say a company needs a new CTO or Engineering Manager and is accepting applications for review. Is there any way a Board of Directors can use quantitative analysis to choose someone from a group of applicants? There’s no real way to convert applicant’s experience, technical ability, workmanship, and interpersonal skills into numerical data to compare against others to make this decision. This ultimately comes down to a judgmental approach such as jury of executive opinion.

Another good example of a purely qualitative problem is when a totally new product such as the iPad is introduced. Forecasting demand for a brand-new product such as this is very difficult due to the lack of any historical sales data (Render, Stair, Hanna, & Hale, 2015). In cases such as these, companies must rely on expert opinion, individual experiences and judgment, and other subjective factors to forecast accurately. The supplemental text of this course does a great job defining qualitative methods used in these scenarios as the Delphi method, jury of executive opinion, sales force composite, and consumer market survey. Each of these have their own benefits which allow companies to forecast accurately without any quantitative analysis. These tools provide immense benefit to clearly weigh empirical data qualitatively.