How to revise a semester of engineering maths in three weeks
Every year, around three weeks before the semester exam, the same message arrives: I have not started, is it too late? It is not. Three weeks is enough for an engineering mathematics paper — but only if you follow these six steps, in this order.
- Stop reading the notes. Start attempting.
The instinct is to read the unit from the beginning and understand everything before attempting questions. With three weeks left, this wastes more time than anything else you could do.
Open a past paper and try a question without looking anything up first. Only go to the notes at the exact point where you get stuck — then find that one method, work two or three examples of it, and go straight back to the paper.
This also tells you, honestly, how much you already know. A student who reads the whole unit first only finds this out in the exam hall. A student who starts with the paper finds out on day one.
- Week one: find the topics that always appear.
Not all units are worth the same. In most engineering maths papers, a handful of topics appear every single year and carry most of the marks. This is not luck — a semester paper is usually built from the same pattern each year, changed only in the numbers or function used.
Lay three or four past papers side by side. Mark every question that repeats, even if the wording has changed. If three of your four papers ask for the same kind of Laplace transform, that is confirmed — put it at the top of your list.
Spend the first week only on those confirmed topics. Five or six topics worked properly beat twelve topics half-covered.
- Week two: sit one full paper, then sort your mistakes.
Sit one complete past paper, timed, with no notes and no phone. It will go badly — that is the point. It is the first honest measurement of where you stand.
Mark it yourself, harshly, the way the examiner will. Sort every mistake into one of three kinds:
- Setup mistake — a wrong limit, a sign carried through, a formula recalled almost correctly. The method was right; the fix is care, not more studying.
- Method mistake — reaching for the wrong technique entirely. That is a real gap — go back to week one’s list.
- Time mistake — a question you could have solved but ran out of room for. That is a pacing problem, fixed by practising faster, not by learning anything new.
Most students find most of their mistakes are setup mistakes, not method mistakes. That means week two is mostly about care and speed, not new mathematics.
- Week three: drill what must be automatic.
The last week is for things that must come without thinking: standard integrals, the transform pairs, the series expansions, the determinant of a three-by-three.
Write them by hand from memory, once a day. Not read — written. Ten minutes each morning, before anything else, is worth more than an hour of reading the textbook.
Add your recurring setup mistakes to this same list. If a sign error keeps happening in the same step, that step needs the same memory drilling as a formula does.
- Do not switch resources in the last week.
A new YouTube video explaining a topic you already understand, watched because it looked clearer, usually costs more time than it saves — it teaches the same method with different notation, and now you have to match that up with what you already know.
Pick one source per topic in week one and stick with it. The only exception: a topic from your method-mistake pile in week two, where a genuinely different explanation can be worth the time.
- The night before: nothing new.
Re-work three questions you have already solved correctly, so you walk in having just succeeded at something. Look back at your error list one last time — by now it knows more about where your marks are going than the textbook does.
If you are in the middle of this and it is not working, bring the paper in to our Sector 12 centre and we will go through it together. That is what the demo class is for.
