Parametric design
Assignment 2
Last updated
Assignment 2
Last updated
The idea of parametric design starts from approximately the 1970’s as an evolution of Computer Aided Design, which originally replicated the human activity of drawing or drafting. Instead of describing directly geometric entities such as points and lines, the key insight was to shift representation to modelling relationships; hence a term often used is also associative or relational geometric modelling.
This paradigm shift allowed for faster design to production cycles reducing the effort of managing and updating design specifications and drawings every time changes were required; design and change are synonymous. The unforeseen opportunity was however in the ability to produce design variations for virtually free. Creating models using parametric methodology defined schematic design spaces capturing notionally infinite design options; single instances from a family of designs.
Unlike the extrapolating, unbounded and often uncontrollable generative design characteristics, parametric design is interpolating, bounded and fundamentally aimed at controlled design variation. The questions arising from the ability to automatically produce multiple options are related to exploring the vast design spaces defined to detect unique instances and defining criteria to compare and select the best options.
Create a parametric design model using the techniques developed. Produce a scalar field family of surfaces and present variation matrices mapping interactions between parameters. Categorize instances along qualitative and quantitative dimensions and after selecting your favourite option, use 3D printing to produce a physical prototype. The theme of this assignment is “Mapping Variations”.
Develop a parametric design describing a family of scalar field surfaces driven by a few key parameters. Follow the steps of the data landscapes session as a guideline. You are free to use any technique desired and motivated to go beyond what is shown as examples in the session notes i.e. do not just reproduce the wavy surface tutorials. Your parametric design schema expresses a multidimensional design space, if we consider each parameter as an independent dimension.
Document design options by creating variation matrices. A variation matrix is a 2D table mapping two parameters; one for columns and one for rows. Each parameter must use three levels i.e. low, centre and high, representing a parameter’s indicative domain. The table cells of this 3 x 3 matrix, contain a picture of the result design.
This mapping allows us to understand the interactions between design parameters and to communicate the available options to third parties; it is a valuable decision support tool for internal and external use.
Summarize the options produced by creating a final design map containing the most interesting instances using two evaluation criteria dimensions, namely qualitative and quantitative for rows and columns respectively.
Quantitative evaluation requires measurable design performance values such as the cost of production, time required for manufacturing, the amount of material used etc. You may use geometric properties that indirectly capture some of those attributes e.g. the height of 3D print determines approximately the amount of time required, the surface area and volume is proportional to cost etc. You may even use online 3D printing services to estimate the cost! It is up to you to define the design metrics.
Qualitative evaluation is categorical instead of numerical. You need to define labels such as low, medium and high, expressing aesthetic, visual and desirability attributes that do not map to numbers in the cardinal sense but into categories in the ordinal sense: you can express how one option is better than another, i.e. sort them from best to worse, but without a specific amount.
The process of 3D printing is fascinating because it is relatively easy to manufacture physical objects. It is an amazing way to convert digital designs, living in the virtual world, into physical artifacts you can hold in your hand. The objective of asking you to choose and print your favourite design is also for motivating you to think ahead of production time; to consider the opportunities and limitations and incorporate them in the design thinking process of part A. All geometries are possible in the virtual world but this is not the case in the physical world. So consider the actual scale and resolution of your design before pressing the print button.
Production of physical objects follows certain requirements expressed as the design specification. Your rapid prototype in plan i.e. top view must be 95mm by 95mm. This is derived by the 3D printer’s allowable working envelope in the XY-plane. The lowest to highest point in the vertical direction shall not exceed 20mm. You may use the bounding box command to evaluate. This constraint is mandated by logistics: we have limited number printers, rapid prototyping is rather slow, the height of print is a key factor of print time and the number of students this year is large.
Additive manufacturing is simple to prepare but it takes significant amount of time in printing per prototype. As FabLab resources and time are finite, we need to be super punctual in scheduling and issuing data for 3D printing.
We ask you to spend Week 04 on design such that by the end of the first session in Week 05, you will have the rapid prototyping files ready.
The teaching assistant will help check your STL files by the end of first session in Week 05
Make sure your file name contains the ID of one student in the group. For example if his/her student ID is “01234567” then name the file “SC01-01234567.stl“.
Organise yourselves in groups of 4 and combine your files in one STL. Book a 3d printing slot and bring your file to FabLab for printing.
Please collect your print when it is ready.
Week 05 is dedicated to preparing your presentation submission while your design is being 3D printed.
Submit a deck of 5 slides documenting the design process and outcomes achieved. Use the linked template below. Additional, information is contained in the template.
Submit your powerpoint to Edimensions. The Edimensions link will be made available on 12th October. The deadline is as follows (variable deadline is due to assignment being handed out on different days depending on cohort)
October 17th (Thursday) 6pm for SC03, SC05, SC09
October 18th (Friday) 6pm for SC01, SC03, SC04, SC06, SC07, SC08, SC10, SC11
October 19th (Saturday) 6pm for SC02
Construct variation matrices by using all pairs of parameter combinations [] using the formula “n! / 2( n – 2 )!”, where “n” is the number of parameters, to compute the number of matrices required. For two parameters you need one matrix; three maps to three; four to six; five to ten; six to fifteen etc. Therefore, it is generally advisable to keep the number of varying parameters to maximum four; keep all others constant.
