Types of Decision-Making Environments & Steps on Quantitative Analysis

Module 3: Types of decision-making environments

There are three basic types of decision-making environments mentioned in Module 3,

• Decision making under certainty - The decision-maker knows with certainty the consequences of every alternative or decision choice

• Decision making under uncertainty - The decision-maker does not know the probabilities of the various outcom

• Decision making under risk - The decision-maker knows the probabilities of the various outcomes

The easiest environment is deciding certainty, when the cause and effect are known and predictable, the risk can reduce to its minimal level. For example, when you roll a dice once, the outcome is one of the numbers within one to six, or if you decide to purchase a laptop for $1000, you will have $1000 less to save in the bank. The decision can be made when the outcome is under the condition of certainty, measurable information The common technique to use under certainty is linear uch as profit) can be expressed as a mathematical function (the value model or objective function) of several variables.

In reality, most of the decision made is in a complex environment that is under uncertainty. The future is unpredictable, and available information might not be accurate or reliable, the decision-maker should not easily make the final decision, rather, they need to make reasonable assumptions and use the most suitable approach to making the decision. There are five possible options: Maximax (optimistic), Maximin (pessimistic), Criterion of realism (Hurwicz), Equally likely (Laplace), Minimax regret. For example, I need to forecast the budget to spend on the laptops for a new hire to meet the HR team's hiring plan, however, due to the current chip supply chain shortage, I need to work with the procurement team to understand the number of the maximin laptop, or minimax laptop can possibility procure, then work with HR team on the hiring plan whether we should slow down on hiring or seek for alternatives.

In decision-making under a risk environment, the approach is to determine the expected monetary value for the alternative, focusing on mathematic approach of possibility.

-----------------------------------------------------------------------------------

Quantitative Analysis | The Steps and Relevant Application

This week, I've chosen to start a discussion about the seven steps of quantitative analysis. According to our text and lectures, they are defined as follows (Zlatev, 2021):

1. Defining the problem.

2. Developing a model.

3. Acquiring input data.

4. Developing a solution.

5. Testing the solution.

6. Analyzing the results.

7. Implementing the results.

There is a cyclical nature to some of the steps, as the first solution developed and tested might not always be the best. In that case, after the solution is tested, the cycle re-begins with the development of a new model.

Furthermore, there are a number of different ways to analyze the results. What-If-Analysis, Sensitivity Analysis, Goal-Seeking Analysis, and Optimization Analysis are all viable methods of determining the efficacy or success of a solution.

When it comes to applying decision-making methods in my current profession, I am fond of using quantitative analysis. As a Sales Operations Analyst who often builds models, forecasts, and makes business-essential decisions, it is all but expected that my decisions are based on copious amounts of data, trials/testing, and analysis. In my experience it is the most holistic approach to decision-making and ensures that biases are minimized.

An example of a time when I had to use quantitative analysis to make a choice was when I built our account tiering system from scratch. I had to determine what variables our organization cared about, what the best combinations of criteria were, and how to automate the process of assigning tiers using Salesforce. After building and testing different combinations of criterion and comparing the expected yields per tier to the desired, I came up with, tested, and implemented a successful tier assignment automation rule. My process, step by step, was as follows:

• The problem? We had no hierarchy of accounts so our reps did not know what to target.

• The ideal model? Assigned tiers to accounts upon creation based on certain criteria. If criteria was changed or updated, the assigned tier would likewise update.

• Acquired input data? Years of sales data, verticals from our ICP, etc.

• Developed solution? An automatic tier assignment rule that used specific, foolproof criteria to distinguish tiers.

• Testing the solution was done by myself and the other members of the sales team for months.

• Analyzing the results was done in a number of different ways but was particularly focused on comparing ideal yields per tier to actual. We also had users test the criteria and see if they agreed with the would be assigned tier.

• Implementing the results came in the form of a successful automation rule which was written by our developers and turned on via the Developer Console in SFDC. Our sales team was trained on how to interpret and action accounts assigned to each tier.

Resources:

• Lectures and Notes, Zlatev, 2021.

• O'Brien, J. (1936-). Chapter 10 - Supporting Decision Making. ExLibris. Retrieved November 15, 2021, from https://bu.leganto.exlibrisgroup.com/discovery/delivery/01BOSU_INST:AlmaGeneralView/12855646900001161? lang=en&viewer ServiceCode=AlmaViewer&loginld=32675749620001161_923LudqHVUbliRr.