Problem statement: First, an undergraduate project aims to give you an opportunity to obtain hands-on experience in design to solve a problem. A high-end project would also have some research element to it. So, you should be very clear about the design problem and about the scientific question to be addressed in the project. Do not have many problems to be solved in an individual final year project, unless it is a group project where different individuals can address different design problems. In essence you should be able to write an email to a friend and explain what you are working on in one sentence. This is your “problem statement”.
Let me take some examples: A problem statement like “Find an energy efficient method to do quality assurance in an already demined landmine field” can lead to a design like this – “Ball that rolls in a mine field”
A problem statement like “Find a solution to avoid neck injuries in women who carry heavy water pails on their heads” can lead to a design like this – “water wheel”
Design approach: I know you are worried about marks. If you solve a problem like the above and show it to the panel without sophisticated math and software, you are not likely to score good marks. If you look closely at above solutions, you might see that the solution required a lot of modeling before the hardware fabrication. You can use design software like Solidworks and simulate dynamics. For instance, in the above water wheel solution, you can model its inertia dynamics and gravity forces and predict the forces experienced by the woman who would be driving it in different slopes and at different speeds. That can help you to decide an upper margin for the capacity of the water drum. First sketch your ideas on a scrapbook, then try to develop very abstract models so that you can write basic equations for them. Then write a simulation code in any language you like (we often use Matlab to simulate dynamic systems), and plot results. For instance, the above ball that rolls in the wind can be modeled like a rimless wheel driven by a random torque give around its hub. You can model wind force by treating the torque as a random variable with a normal distribution around some expected value. Then, you can check the distribution of collision forces for different hub masses and wind conditions. This will allow you to design a ball that can generate enough collision forces to detonate a landmine buried at a reasonable depth, and to know the balls limitations too.
This is where I can guide you if you meet me (your project advisor) frequently. Once a week is recommended.
Documentation: You must not waste any of your intermediate designs on the way to a final solution. Some students show only the thing that worked in their final report. Present all results including your decision tree where you took one choice over the other at different milestones of the design process. Again, meet me at every milestone. I can help you to take a good choice given several options. I recommend you to use a scrapbook as well as saving your work in backup folders.
Dissertation writing: You are given marks for this. No matter how great your engineering was, you could lose marks if you write a bad report. Your documentation can help here. I highly recommend you to get yourself familiarized with Latex dissertation writing from the beginning of your project because for me, a dissertation that involves math looks more professional if it is written in Latex. Moreover, in Latex, you don’t have to worry about formatting. You just use a style file and the compiler will take care of formatting allowing you to focus more on the content. The best way to get used to it would be to keep your notes and documentation in Latex. Learn to write equations, how to insert a figure, how to write a pseudocode/algorithm in Latex without waiting till the last moment.
Organize your results in a methodical way so that your intermediate results can show the lessons you have learned on your way to the final result. Please meet me even during dissertation writing. Please be assured, all those who took these advices scored very high marks (> 80%) and even published papers in conferences, and then got good jobs!