Inverse Material Design Project
The capstone — pose and solve an inverse design problem for a real Chinese food material, using the course's models run backward.
Due: 26 May 2027 · Weight: 35%
The brief#
Choose a real Chinese food material and pose an inverse design problem for it — a specific, measurable target outcome you have not simply been given — then solve it by inverting one or more of the course’s forward models, and report what confidence your answer actually deserves.
This is the course’s synthesis project: it draws on whichever governing models fit your chosen material and target, not on any one module alone. A strong project states its target with a number and a tolerance (following Module 12’s framing), is explicit about which equation each free variable acts on, and treats the inverted answer as a hypothesis to be tested against real data rather than a finished specification — the same caution the course’s own module pages apply to every model they present.
What you submit#
A single PDF (2,500–3,500 words excluding references and appendices), structured as: the target outcome and why it is measurable, the forward model(s) selected and why they apply to your material, the inversion itself with working shown, and a closing section stating the confidence and limitations of the result. Supporting calculations may go in an appendix.
The final project is judged as a whole rather than against a weighted checklist — see the marking statement above — because the scope of a good inverse-design problem varies more than a fixed criteria table can fairly capture across different chosen materials.
The spec
The brief poses the problem and leaves room for a response. The spec is the fixed part: what a marker checks. Some lines can be checked mechanically; the rest call for judgement.
- submitted by the deadline as a single PDF, 2,500–3,500 words excluding references and appendices
- states one specific, measurable target outcome with a tolerance, not a texture word alone
- names the forward model(s) inverted and the free variables acted on, tracing back to specific equations
How it is marked
Judged as a whole against the brief, with particular weight on whether the stated target outcome is genuinely measurable, whether the chosen forward models are appropriate to invert for it, and whether the submission is honest about where its inverted answer is a tested hypothesis rather than a guaranteed specification.