Turn a business choice into a model you can argue with, solve and defend
A modelling course for anyone who has to choose between alternatives and wants the choice to stand up to scrutiny. It works through the standard toolkit of quantitative decision making: framing a decision, formulating and solving optimisation models, testing how sensitive an answer is, learning relationships from data, forecasting a series forward, and handling uncertainty by simulation and decision trees. Every model is built the way practitioners build it, in a spreadsheet, with the algebra alongside. Nano College keeps the ground the textbook covers and the rigour it insists on, and leaves out the padding; each piece stands on its own, so you can open the model type you need today instead of reading a semester in order.
Coming soon
Why a decision worth modelling is a decision worth getting right, and what an optimisation problem looks like before you solve it.
Why a decision worth modelling is a decision worth getting right, and what an optimisation problem looks like before you solve it.
Turning a formulation into working cells, driving the solver, and reading what the sensitivity report is really saying.
Turning a formulation into working cells, driving the solver, and reading what the sensitivity report is really saying.
Optimisation where the structure is a graph, or where the answer has to be a whole number — and the two are more alike than they look.
Optimisation where the structure is a graph, or where the answer has to be a whole number — and the two are more alike than they look.
What to do when there is no single number to maximise, and what changes when the objective or the constraints stop being straight lines.
What to do when there is no single number to maximise, and what changes when the objective or the constraints stop being straight lines.
Learning a relationship from data: fitting it, trusting it, and then letting algorithms search for structure you did not specify in advance.
Learning a relationship from data: fitting it, trusting it, and then letting algorithms search for structure you did not specify in advance.
A series carries its own structure — level, trend and season — and each forecasting method is a bet about which of those is stable.
A series carries its own structure — level, trend and season — and each forecasting method is a bet about which of those is stable.
When the arithmetic is too tangled for a formula, sample it; and when customers queue, work out what the wait costs.
When the arithmetic is too tangled for a formula, sample it; and when customers queue, work out what the wait costs.
Choosing between alternatives you cannot fully control, then committing to a plan of activities and a finish date you have to defend.
Choosing between alternatives you cannot fully control, then committing to a plan of activities and a finish date you have to defend.
“The rare modelling book whose first lesson is that solving is the easy part.”
Most treatments rush to the software. This one spends its opening units on framing the decision and judging what a model can honestly claim, then builds every method out in full. The sequencing rewards patience: sensitivity analysis lands only after you have felt an optimum go wrong.
Nano College takes the standard university textbook for a course and turns it into short, self-contained pieces you can actually finish. We keep the ground the textbook covers and the rigour it insists on; what we leave out is the padding that only exists to fill a semester.