Motivations
Management is the attainment of organizational goals in an effective and efficient manners through planning, organizing, leading, and controlling organizational resources.
Daft R. L. (2014) Management
When you run a company or an organization, you want to do something, that is your goal, but you have limited resources, you want to assign or allocate these resources to several different kinds of activities or operations. The resources may include: limited budget, time, number of employees, amount of shell spaces, etc. You want to do some research about how to do it, that is Operations Research.
Operations Research
Operations Research is the methodology to allocate the available resources to the various activities in a way that is most effective for the organization as a whole. It is applied to problems that concern how to conduct and coordinate the operations (activities) within an organization.
Hillier and Lieberman
Operations Management is your target, and Operations Research (OR) is the methodology. But OR is also applied to marketing, accounting, human resource management, information systems, and finance. OR basically helps you make decisions. It is a tool to support decision making. Typically it does not make decisions for you. It make suggestions to the decision makers.
Business Analytics
Business Analytics has a tight connection with Operations Research. Basically, BA has 3 levels:
| Descriptive analytics (statistics) | What happened | With historical data |
| Predictive analytics (machine learning) | What will happen | With historical data With prediction model |
| Prescriptive analytics (operations research) | What should we do | With historical data With prediction model With decision model |
A typical process of decision making using Business Analytics consists of 2 steps:
- Data analysis – collect information and understand the problem.
- Descriptive
- Predictive
- Operations research – allocate resources and solve the problem.
- Prescriptive
All kinds of machine learning modules are doing predictions. With those information, Operation Research tries to arrange your resources, and help make the best decision.
Process of Operations Research
- Collecting data
- * Define the problem, go to 1. if data is not sufficient
- ^ Formulate a model
- ^ Solve the model, go to 3. if model is not good
- * Interpret results and make suggestions
The steps with * above are critical and more about application, but steps with ^ are more fundamental and theoretical.
Analytics can not do everything
| Advantage | Use math to describe a problem in a precise and concise way. Facilitate computers to solve a problem. |
| Disadvantage | Some of the problems are too hard to be formulated. Some critical information is missing. |
Mathematical Programming
The reason why we need a model is that we need a precise description about the problem so that computer can understand. When the problem is complicated enough, simple rule may not work (ie, fail to find an optimal solution), but with a model, computers may help find a better solution. For a mathematical programming model or formulation, we need to define:
- decision variable – what may we determine
- objective function – what do you want
- constraints – what are the limitations
The model / formulation can be presented in a compact and more abstract format. This abstraction saves space and is easy to generalize to other applications.
There are various types of mathematical programs, for example Linear Programming and Integer Programming. They look similar but they are actually different, suitable for different real problems.
| Linear Programming | decision variable falls into an interval: 0 ≦ x ≦ 1 |
| Integer Programming | decision variable must be an discrete integer: x ∈ {0, 1} |
My Certificate
For more on Operations Research, please refer to the wonderful course here https://www.coursera.org/learn/operations-research-modeling
Related Quick Recap
I am Kesler Zhu, thank you for visiting my website. Checkout more course reviews at https://KZHU.ai
