So, you're thinking about becoming a simulation modeler and want to find out more. Well, you've come to the right place! If you have already begun your simulation modeling journey, you might want to check out this guide on how to go from good to great in simulation modeling.
In this article, we will detail what a simulation model developer does, what you need to know to become one, and where to find learning resources.
Professional simulation modeling
Around the world, computer-based simulation modeling is a relatively new technology in business. Companies such as Siemens, Microsoft, PwC, Lockheed Martin, Pfizer, Alstom, McDonald’s, Airbus, DHL, Walmart, BMW, and many more now use simulation modeling as part of their daily workflows. Why do they choose simulation?
With its help, companies plan changes, optimize processes, test hypotheses, and generally improve performance. In addition, modern business processes and systems are becoming more complex, and traditional methods of static modeling in Excel are no longer adequate.
Simulation modeling is the reconstruction of a real system using a computer program, where the model reflects the structure and behavior of the system over time. In other words, this modeling technology differs from the rest by taking into account the dynamics of the system.
What do simulation modelers do?
Modeling for business is slightly different from modeling for engineering and scientific projects, such as aerodynamics, chemistry, meteorology, and mechanics. This article will focus only on the developers of simulation models for solving applied business problems.
Simply put, a simulation modeler builds models of real objects and systems.
Usually, they have higher education in the sciences and programming skills. They know how to design data structures and write algorithms for working with them, they know the basics of simulation, statistics, and probability theory. At the same time, they are familiar with modeling the dynamics of continuous media and graphic design, and ...
...and, okay, that's in an ideal world. In reality, the requirements for simulation modeling specialists are more flexible.
From our experience, people enter this profession in one of two ways:
- as a person with a technical or economic education, who is already remotely familiar with technology, and who decides to delve into this area;
- as a specialist in their subject area, such as mining, manufacturing, or logistics, and who studies simulation modeling in order to more effectively solve problems in their industry.
What do you need to know to become a specialist in simulation modeling?
1. Willingness to learn, flexibility, and personal motivation
Of course, knowledge of statistics, mathematics, and algorithms is necessary, and a technical education will greatly help in working with models. But, if you do not have this knowledge, you can take specialized courses.
In addition, you need to learn what simulation is, what problems it solves, what alternatives there are (linear programming, optimization) and what they are used for, and to get acquainted with the tools that are on the market.
The sooner you start trying to build models, the sooner you will learn. When working with models, you will inevitably need to work through difficulties. By doing so, you will be well on the road to becoming a professional..
Start with the basics. Take a modeling course or work through the self-study materials, download the free version of AnyLogic and try building basic models with step-by-step instructions. Gradually move on to real projects and see them thorough to completion.
To become a pro, try new approaches to solving the same problem, try to understand the logic of models created by other specialists (if source files are available), and test the different possibilities available in the modeling tool you are using.
2. “This is not what we wanted”, or the importance of communication
If you have chosen the path of the simulation modeler for yourself, be prepared for the fact that you will need to talk. A lot. With many different people.
Commonly, there is a customer in the project who wants to solve a business problem, and you offer them a solution using simulation. It is important to remember that you are not only solving a technical problem (how to model), but also a business problem (will simulation help solve the problem?).
In practice, it often turns out that what is written in the terms of reference, spoken at the meetings of team members, and the result expected by the customer's management are three very different things.
Before you start working on a model, it is important to clearly articulate the goals and define the end result of the project so that your expectations (as well as your team) and the customer's expectations match as closely as possible. In other words, you need to learn how to manage the expectations of the stakeholders in the project and maintain communication throughout.
3. The art of simplifying (yes, it’s an art)
"The choice of abstraction level and modeling method is a central issue in building a model. What assumptions can we make? Where are the boundaries of my model, that is, what do we consider given and external, and what do we model? This is the realm of art, experience and intuition, not science and technology."
Imagine that you have some kind of complex system, such as a factory with workshops, people, equipment, processes, etc. “Probably, the more detailed I model this system, the more reliable the result will be,” or so the novice model developer thinks. It is a typical mistake.
Choose a level of abstraction for a model only so that it will solve the problem at hand, no need to detail more than is necessary. Follow the principle of Occam's Razor: if you can do without a part, don't include it in the model. The level of abstraction also determines which modeling methods are best suited in a particular situation. The main methods are discrete-event, agent-based and system-dynamic.
Who can be a simulation modeler?
Simulation may be your main activity or an additional area of expertise. Regardless of where it fits in your work, knowledge of its principles and capabilities, as well as the ability to create and apply models, is a requirement for modern engineers, managers, business analysts and business consultants.
Managers, for example, use simulation and a systems approach to manage complex projects with many interconnections. In this way, they can prevent project disruption or cost overruns. They can also safely test their hypotheses and use models to support their proposals.
Simulation modeling is also used by logistics professionals to design new and improve existing supply chains. Modeling helps to evaluate the effectiveness of logistics processes, develop strategies and solutions for supply chain management.
Where is simulation modeling taught?
Simulation modeling is studied in undergraduate and graduate programs at a number of major universities, including:
- York University – Disaster Simulation, an agent-based approach
- MIT edX – MicroMasters in Supply Chain Management
- Singapore University of Technology and Design – 40.015 Simulation Modelling and Analysis
- University of Saskatchewan – Agent-based Modeling for Health Policy
Simulation modeling is a relatively young technology for business, but it is already in demand among large companies. It is used by top managers, logisticians, business consultants, and analysts to solve problems in their fields of activity.
If you have some of the knowledge and skills listed in this article, a great desire to master simulation modeling and work with such projects, we advise you to start learning with the materials on our website:
- Workshops and Seminars
- Getting Started resource collection
- The free Personal Learning Edition of AnyLogic
Along the way, you will have questions and difficulties. Don't let that scare you. Remember that there are many people in the world who, just like you, are learning how to build simulation models. Post your questions on StackOverflow (How to win at StackOverflow) and we'll be sure to help you find the answers.