How to Optimize Engineering Designing?

Engineering drawing is a graphical representation of a concept, an idea, or an entity that actually or can potentially exist in life. It allows efficient and effective communication among engineers and other technicians. It communicates all the information clearly and concisely to engineers. An engineering drawing contains notes, specifications, and dimensions. It is a preferred method of drafting in civil engineering, architecture, and civil engineering. It is a universal language.

Engineering drawing is better than a written plan as it develops the skills to produce complex ideas into simple engineering drawings and sketches. It develops the ability in drawing data and information from diagrams and calculation sheets to develop drawings for fabricators, installers, and manufacturers. It helps in creating knowledge of plant and equipment layout. It is also useful in understanding complex construction and manufacturing drawings used in the industry. In electrical engineering, it is used for constructing wiring diagrams of homes and industries. It is also used for circuit diagrams and as electrical installation drawings. In civil engineering, it is done to create foundation drawings and other architectural plans. It is highly recommended while creating engineering products in order to mitigate any errors in the design process.

Around the world researchers and engineers face many hurdles in creating optimized engineering designs. As computers have taken the world by storm, optimization has become a part of computer aided engineering. It is useful in achieving design objectives by either minimizing the cost of production or by maximizing the efficiency of production. It is a technique to achieve the optimum solution after analyzing all the alternatives. It involves a comparison of varied chosen designs and thereafter selecting the optimum industrial design.

Engineers and researchers have to deal with the adjustment of many variables to achieve varied objectives. An optimum design can be achieved with a mix of modeling, experience, and correct judgment.

Today, there are generally two types of design optimization algorithms. One of the types of algorithm is deterministic and uses specific procedures to move from one solution to another solution. These types of solutions have been successfully used for quite some time to design many engineering problems. The second type of optimization algorithms uses probabilistic transition rules and is random in nature. These algorithms are slowly becoming part of an engineer’s world. Though, both of these algorithms’ use has advantages and disadvantages of various forms due to complex engineering designs.

An essential part of the optimal design process is the transformation of the design problem in a mathematical format. A mathematical format is generally a combination of variables and functions. These variables and functions can be analyzed for engineering optimization. Engineering models are created to carry out engineering optimization. Analysis Variables are the inputs that are given by the user to make calculations of a particular design. These include boundary conditions and material designs, etc. The analysis model is evaluated after setting all the parameters of the variables. The analysis model estimates the output which is also known as analysis functions. The representation of these analysis functions signals the optimization of the design.

The analysis software computes the functions after analyzing the designer’s input for analysis variables. The software does not take decide the optimization of the design but has to rely on the judgment of the user. The designer gives a set of input parameters to the software for evaluation which in turn gives the output for further examination. Designers utilize their knowledge and experience to evaluate the model many times before reaching the best solution. Such a process is called optimization by trial and error.

A computer-based optimization approach is now increasingly being utilized to analyze the model. It is also useful for researching a better design. Here, the computer help in optimizing the result by taking the objectives of the design problem from a designer. Designers try to tackle complicated designs by following different approaches. Some of these optimization approaches are:

Unconstrained Optimization:

Unconstrained optimization focuses on the problem of minimizing the objective function that depends on real variables. These problems are due to indirect reformulations of constrained optimization designs. It is usually advised to replace the constraints of an optimization problem in the objective function before solving for unconstrained problems.

Discrete Variable Optimization:

Many problems that arise while designing are in models where design variables are selected from a set of discrete values. These discrete variables can be of beams, springs, fasteners, pipes, etc. One approach for solving the discrete problems is ‘Branch and Bound’. It focuses on building a tree structure and thereby estimating the objective function.

Genetic and Evolutionary Optimization:

Evolutionary optimization focuses on the process of optimization in nature. It copies the processes of the biological species that are undertaken for the survival of the fittest. It helps the designers by solving complex problems with discrete value and large number of design variables. It is also helpful with multiple local minima, maxima, and saddle points.

Optimization of engineering designs is robust and powerful but a careful vigil must be carried out while designing the project. The engineer should validate the engineering model carefully as an inaccurate model can lead to a waste of time and work. Most of the algorithms are preset and usually have loopholes. The algorithms should be custom made as per the design requirements. It is the responsibility of the engineer to select the optimized design as algorithms will give results only on the basis of sets of inputs. Quantitative optimization and quality of concept selection are required to achieve the desired optimization. System optimization should be given priority along with the individual components. Many times, designers come through conflicting design objectives hence sometimes engineers must also consider the designs non-quantitatively.

Optimized designs are prone to variation due to active or binding constraints. Researchers and engineers should not only focus on optimized designs but also for robust designs that can tolerate the variation. The feasibility of the robustness of a design depends on the toleration variability and functional property. Hence, an optimized design should also be a robust design in order to be the best alternative among various designs.

Value engineering in the modern era requires developing regular comparable data so the results are frequently accessible and readily used. It caters to bring better decision making and improves the quality of the product in the long term.