Revision notes

Chapter 12 Linear Programming: Foundation

The key points, rules to remember, tables and diagrams from all 5 lessons, on one page for quick revision. To keep a copy, use your browser's Print and choose "Save as PDF".

Lesson 1

Optimisation problems and the language of linear programming

Terms used in linear programming
TermMeaningIn the tailor problem
Decision variablesThe unknown quantities to be decided; always non-negativex shirts, y kurtas
Objective functionThe linear function Z = ax + by to be maximised or minimisedZ = 100x + 180y
ConstraintsLinear inequalities or equations that the variables must satisfy2x + 3y ≤ 60, x + 2y ≤ 35
Non-negative restrictionsThe conditions x ≥ 0, y ≥ 0Cannot make a negative number of garments
Optimal valueThe maximum or minimum value of ZTo be found

Parts of a linear programming problem

  • Decision variables
    • x, y ≥ 0
    • What we choose
  • Objective function
    • Z = ax + by
    • Maximise profit or minimise cost
  • Constraints
    • Linear inequalities such as 2x + 3y ≤ 60
    • Come from limited resources or requirements

An LPP: optimise a linear objective function Z = ax + by subject to linear constraints and x ≥ 0, y ≥ 0.

  • An optimisation problem seeks the best value of a quantity under limits.
  • In an LPP, the objective function and all constraints are linear, and the variables are non-negative.
  • Decision variables are the quantities you choose; constraints come from limited resources or minimum requirements.
  • 'At most' becomes ≤; 'at least' becomes ≥.
  • Trying plans by hand shows the need for a systematic method, developed in the next lessons.
Read the full lesson

Lesson 2

Formulating linear programming problems

From words to an LPP

  1. Identify what is to be decided; name the decision variables x and y with units
  2. Identify the goal: maximise (profit, output) or minimise (cost, time)
  3. Write the objective function Z = ax + by
  4. Turn each limit or requirement into a linear inequality
  5. Add x ≥ 0, y ≥ 0
Data for worked example 1
ResourcePer toy APer toy BAvailable
Cutting (hours)32120
Painting (hours)1280
Profit (₹)2540To be maximised
Translating common phrases
PhraseInequality
At most 30 itemsx + y ≤ 30
At least 20 g of protein4x + 2y ≥ 20
At least as many lamps as fansy ≥ x
The number of A is at least twice the number of Bx ≥ 2y
Exactly 10 units in totalx + y = 10
Cannot exceed 120 hours3x + 2y ≤ 120

One constraint per limited resource or requirement; 'at most / cannot exceed' gives ≤, 'at least / not less than' gives ≥; never forget x ≥ 0, y ≥ 0.

  • Define the decision variables clearly, with units.
  • Profit or output problems are usually maximised; cost problems minimised.
  • A data table gives each constraint as a row.
  • Simplify constraints by dividing by common factors, but keep the inequality sign the right way round.
Read the full lesson

Lesson 3

The feasible region and its corner points

Testing points against the tailor constraints
Point2x + 3y ≤ 60?x + 2y ≤ 35?Feasible?
(10, 10)50 ✓30 ✓Yes
(15, 10)60 ✓35 ✓Yes (a corner)
(20, 10)70 ✗40 ✗No
(0, 18)54 ✓36 ✗No

Finding the feasible region and its corners

  1. Draw each boundary line using its intercepts
  2. Shade the correct side of each (test the origin)
  3. Take the common region in the first quadrant
  4. Mark corners; solve pairs of line equations for their coordinates
  5. Note whether the region is bounded or unbounded

Bounded and unbounded feasible regions

  • Bounded
    • Fits inside some circle
    • A closed polygon
    • Typical of ≤ constraints
    • Example: the tailor problem
  • Unbounded
    • Extends indefinitely
    • Not closed on one side
    • Typical of ≥ constraints
    • Example: the canteen problem
  • Each linear constraint is a half-plane; test the origin to choose the side.
  • The feasible region is the common part, inside the first quadrant.
  • Corner points are intersections of boundary lines that satisfy all constraints.
  • Bounded regions fit in a circle; unbounded regions extend indefinitely.
  • The feasible region is always convex.
Read the full lesson

Lesson 4

The corner point method for bounded regions

Bounded feasible region: evaluate Z at every corner point. The largest value is the maximum and the smallest is the minimum.

The corner point method (bounded region)

  1. Draw the feasible region
  2. Find the coordinates of all corner points
  3. Evaluate Z = ax + by at each corner
  4. Largest value M = maximum; smallest value m = minimum
Z = 100x + 180y at the corners
Corner pointValue of Z (₹)
(0, 0)0
(30, 0)3000
(15, 10)3300 (maximum)
(0, 17.5)3150
Z = 3x + 5y at the corners
Corner pointValue of Z
(4, 0)12
(4, 2)22
(0, 6)30 (maximum)
(0, 2)10 (minimum)
  • Theorem 1: an optimal value, when it exists, occurs at a corner point.
  • Theorem 2: on a bounded region, Z has both a maximum and a minimum, at corners.
  • Corner point method: find corners, tabulate Z, read off the largest and smallest values.
  • The origin is a corner only when it satisfies every constraint.
Read the full lesson

Lesson 5

Unbounded regions, multiple optimal solutions and infeasible problems

Checking an unbounded region

  1. Evaluate Z at all corners; note M (largest) and m (smallest)
  2. For a minimum: draw ax + by = m and look at the side ax + by < m
  3. If that open half-plane misses R, m is the minimum
  4. If it meets R, there is no minimum (similarly for M and a maximum)
Possible outcomes of an LPP
SituationWhat the corner point method gives
Bounded regionBoth maximum and minimum exist, at corners
Unbounded regionUse the open half-plane test; an optimum may or may not exist
Two corners tie for the optimumEvery point of the edge joining them is optimal
Contradictory constraintsNo feasible region, no solution
  • On an unbounded region, test the open half-plane beyond the best corner value before calling it optimal.
  • Minimising a cost with positive coefficients over an unbounded region of ≥ constraints usually has a minimum but no maximum.
  • Two optimal corners mean infinitely many optimal solutions along the edge between them.
  • Contradictory constraints give an empty feasible region and no solution.
  • The feasible region is convex and the optimum, when it exists, is at a corner.
Read the full lesson

Privacy · Terms · Refunds · Delivery