Interpreting the optimum in context
Interpreting the solution in context
- The maths gives a vertex; the answer is a sentence about the situation.
- State the optimal values of the decision variables with their meaning and units ("make 3 tables and 4 chairs").
- Give the optimal value of the objective too ("for a maximum profit of $240").
- Check it is sensible — non-negative amounts, within every constraint, and reasonable for the context.
Integer-only solutions
- Many contexts only allow whole numbers — you cannot make tables or pack of a tray.
- The graphical optimum may land on a vertex with fractional coordinates.
- When that happens:
- Do not simply round — a rounded point can fall outside the feasible region.
- Test the whole-number points near the optimal vertex that are still feasible.
- Choose the feasible integer point that gives the best objective value.
Solutions on an edge (multiple optima)
- Usually one vertex wins. But if the objective line is parallel to a binding constraint edge, then every point along that edge gives the same optimal value.
- This is a multiple-optima (or "infinitely many solutions") case.
- How to spot it: two adjacent vertices give the equal best objective value.
- How to report it: state that any point on that edge is optimal — often giving the business useful flexibility in what to make.
Checking your answer makes sense
- Substitute the optimal point back into every constraint — it must satisfy them all.
- Confirm the objective value matches your vertex test.
- Re-read the question — answer exactly what was asked (a plan, a profit, or both), with units.
A Dunedin bakery makes batches of loaves and batches of pies (whole batches only). Oven time gives and flour gives , with . Profit is (hundreds of dollars). Find the best whole-batch plan.
Step 1 — Find the graphical vertices
Corners: , (from ), (from ), and where the two lines cross.
Solve and . Subtracting: , so and , giving .
Step 2 — Test the vertices
The graphical maximum is at — but batches is not allowed.
Step 3 — Test feasible whole-number points near
- : ✓, ✓, so feasible. .
- : ✓, ✓, so feasible. .
- : feasible, .
Step 4 — Choose the best feasible integer plan
The largest feasible whole-batch profit is at .
Test yourself
Practice by grade
One question each at Achieved, Merit and Excellence. Have a go, then compare with the model answer.
A linear programming optimum is at the vertex with profit $240, where = tables and = chairs. Write a full sentence interpreting this result in context.
A graphical optimum is at for a problem where and must be whole numbers, with objective and constraints and . Find the best feasible integer solution.
A feasible region has vertices , , and . The objective is . Show that this problem has multiple optimal solutions, identify them all, and explain what this means for the business.