Knowledge Base

Product Modelling

Product Modelling and Modelling Software

Intuitively, everyone from the engineering stream knows what modelling and simulation mean. However, when it comes to precise definition of the terms and more importantly, the role they play in product design and development is something that most people cannot define clearly. In this article, we will try to explore in detail modelling, and various modelling software tools that help engineers and designers come up with better products. While modelling and simulation are almost invariably interlinked in product development, we will discuss them separately. In this article, we will deal with modelling; simulation will be discussed in a subsequent article.

Engineering Modelling
A model is a presentation or abstraction of the product or the system that is digitally developed. They can be visual or mathematical. A miniature car that shows all the features of the real version, a skeleton that shows various parts of the human body all examples of visual models. On the other hand, an atom that shows the path of the electrons revolving around the nucleus and a video depicting computation of gravitational waves are examples of scientific models developed using underlying laws of mathematics. One of the most important uses of modelling is in mechanical engineering and product development industry, for example the automobile industry. Crash tests that involve structural mechanics, deep drawing that uses structural optimization, vehicle aerodynamics or air conditioning involving fluid mechanics, fuel injection which involves combustion engineering, vehicle dynamics for optimal control, or sensors and actuators - the automobile industry is one of the most important fields that uses large scale modelling. Be what may and for whichever field, models emulate some kind of a phenomenon or system that make it easier for humans to understand how they work. Broadly speaking, a model is a simplified representation of reality (or what will be reality in case of products and systems) that makes it easy for humans to gain an understanding of how all the individual components function as an unit.

2D and 3D Modelling
Earlier, initial modelling was done manually using sketch boards; today primary models are generated using sophisticated 2D and 3D CAD software. 2D models involve only width and height of an object; they do not involve the depth. 3D modelling adds the perception of depth to the model by adding one more dimension. While most designers use 3D CAD tools for drawing, there are instances where simple width and height suffice. Moreover, a 2D drawing can be created or modified quickly, without making changes to a part or assembly document.

There are three primary types of 3D modelling:

  • Polygonal: this uses 3D shapes such a as cubes or spheres to create models.
  • Curve: this kind of 3D modelling software is based on precise mathematics. It is primarily used where real-life models need to be built later, using 3D printers.
  • Digital sculpting: Relatively a new kid on the block, digital sculpting allows users to literally pull, grab, poke and slice models like clay.

It is important to note that the word ‘model’ is widely used in many walks of life. 2D and 3D CAD software is used to create initial models, mostly from the inputs of product designers. In the context of this article however, when we talk about model and modelling, we mean mathematical modelling or parametric designs as applied to product development. What is the difference? Well, product modelling is much complex than designing in 2D and 3D as it deals with engineering aspects of the design. The objective of product modelling is to develop objects and systems that sustain in the environment for which they are created. Let us take an example. A college going student may use 2D / 3D software to design the model of a drone, but to turn that model into a working prototype (and maybe a successful product later on) requires engineering knowledge. To turn the drone design into a prototype, product engineers evaluate the drag it faces, the wing span needed to keep it afloat, material to choose so that the drone is lightweight yet sturdy, crash test in case it falls to the ground or encounters an obstacle, the optimum camera weight, and so on. Product modelling thus turns a design model into a working model.

Modelling is a very important pre-requisite of successful product development. Before any product or a system can be developed, it is necessary to consider all aspects of design and engineering requirements. Modelling the user interfaces and making scenarios for typical users (use cases) of the system helps to refine the needed transformation processes.A model must have a clear purpose, and must focus on solving a particular problem. If tomorrow, someone says they have modelled 'weather', they are most likely misleading, as ‘weather modelling’ is a very vague term. A better idea would be to model a particular aspect of weather, say 'cyclones in Indian ocean'.

Before starting product development using a modelling system, the design engineers to answer four questions:

  1. Purpose of Product
  2. What properties need to be modelled
  3. Which tools will be used to model
  4. How to objectively assess the quality of the model after it has been developed

Since this is an overview of modelling, we will not delve into the details of these questions; suffice it to know that answers to these three questions allow proper model development. As an aside, modelling and simulation takes about 80% time of the product evolvement process. Simply put, it is an arduous task developing an object / system than manufacturing it.

Industrial Design
The creation of a new product takes in many levels of activity and skills. The first and foremost of these are the skills of design engineers and industrial designers. A considerable part of the work of design engineers and industrial designers consists of formulation of suggestions of shapes and their modelling, and simulating and appraising the various possibilities. Industrial design is the study of form and function, designing the relationship between objects, humans, and spaces. It is a professional service that optimizes function, value, and appearance for the mutual benefit of both user and manufacturer. Industrial design is present in all walks of life – from your mobile phone to the vehicle you drive, and all the gadgets and appliances you use every day. Industrial designers put in a lot many efforts in delivering a product that is functional as well as aesthetic, and modelling is an integral part of this process.

Parametric Design
One of the design challenges when designing with digital models is how to evaluate the relation between a form and its performance. Parametric design is a modelling approach that generates geometric solutions based on a family of related parameters. It revitalizes the design world's interest in iterations because it offers greater variations through a set of controlled mathematical parameters. Designers can parametrically modify the digital model and it emphasizes on the geometric relations between parameters and forms and uses variables and algorithms to quantify such geometric relations. Parametric design software like PTC Creo provides designers with a series of tools to evaluate ideas in the early design phase.

Types of Modelling
There are various types of criteria on which modelling can be classified. One of the most popular methods of classifying modelling is the white box, black box and the gray box.

White Box Model: A white box model is based on the assumption that the process / system can be described entirely from physical insight and prior knowledge. In other words, white box models can be described by mathematical and logical equations, and logical relationships.

Black-Box Model: Black-box models are based on experimental data in which the parameters must all be estimated from measurements of inputs and outputs, using an estimation procedure. In essence, black box models reflect the fact that no knowledge of the process is used. It is useful when a designer’s primary interest is in fitting the data regardless of a particular mathematical structure of the model.

Gray-Box Model: As the name implies, gray box modelling is based on some prior knowledge and some empirical data. It is useful when you know the relationships between variables, constraints on model behaviour, or explicit equations representing system dynamics.

Another important way of classifying a model is whether it is linear or non-linear.

Linear models are used to study how a quantitative variable depends on one or more predictors or explanatory variables. The predictors themselves may be quantitative or qualitative. They assume that each time sample is independent of the next; however a linear model is not restricted to a straight line or its analogue in higher dimensionality.

Nonlinear modelling is empirical or semi-empirical modelling which takes at least some nonlinearities into account.Nonlinear models have more flexibility in capturing complex phenomena than linear models of similar orders.

Let’s make this simple. Suppose a person wants to commute from point A to point B. Linear models assumes the path is straight; nonlinear models assume the path has some turns, signals, diversions, etc. Nonlinear modelling emulates real life in a better fashion, but is more difficult to model than the linear route. As a norm, it is better to experiment with linear modelling first. If that does not produce the desired output, you need to opt for nonlinear modelling.

Software for Engineering Modelling
Every product design starts with concept development, followed by parametric design generation or 3D modelling. In 3D modelling, product designs are developed in greater detail while analysing their manufacturability.

‘Design for Manufacturing’ is an important part of 3D modelling & product design, ‘Design for prototyping or Additive Manufacturing’ helps companies create designs that can be developed into physical models using Additive Manufacturing technologies for better visualization, validation or End-use depending upon its applications.

In the early days of CAD and CAE, there was a limited choice of engineering software. This was primarily because the computing power was limited, and high end machines were very expensive. As the power of motherboard and chips grew, it benefitted both CAD software makers and users. Developers could add new features without worrying too much about the processing power, and users could use these features without worrying about the processing speed. While there are various options for 2D and 3D drawing of objects, there are only a few modelling and simulation software that have proven their mettle. Altair leads the pack with offerings like Inspire™ for concept design. Altair’s Inspire tool can help designers optimize their concept designs using basic structural analysis. In general, Altair is accepted as the industry leader in providing engineering modelling software in India and other countries.

Modelling for product development requires complex mathematical calculations based on various engineering properties. Special modelling software (for example from Altair) is needed to transform the complex mathematical equations into a successful prototype.