Example 1
Question: Maximise \( Z = 2x + 3y \) subject to the constraints \( x - y \le -2 \), \( -x + y \le 1 \), \( x \ge 0 \), \( y \ge 0 \). Comment on whether a maximum value exists.
Solution: We first find the feasible region by graphing all constraints. Constraint 1: \( x - y \le -2 \), i.e. \( y \ge x + 2 \). The boundary line is \( y = x + 2 \). Since \( x \ge 0, y \ge 0 \), this line passes through \( (0, 2) \) and \( (-2, 0) \) (but \( x \ge 0 \) so we only use the part with \( x \ge 0 \)). The region satisfying \( y \ge x + 2 \) with \( x \ge 0 \) lies above this line. Constraint 2: \( -x + y \le 1 \), i.e. \( y \le x + 1 \). The boundary line is \( y = x + 1 \). It passes through \( (0, 1) \) and \( (-1, 0) \). With \( x \ge 0 \), the region satisfying \( y \le x + 1 \) lies below or on this line. Now we check whether both constraints can be satisfied simultaneously (with \( x \ge 0, y \ge 0 \)). From Constraint 1: \( y \ge x + 2 \). From Constraint 2: \( y \le x + 1 \). For both to hold: \( x + 2 \le y \le x + 1 \), which requires \( x + 2 \le x + 1 \), i.e. \( 2 \le 1 \). This is a contradiction. Therefore, there is NO point \( (x, y) \) that satisfies all the given constraints simultaneously. The feasible region is empty. Since the feasible region is empty, the Linear Programming Problem has no feasible solution, and hence Z has no maximum value.
Answer: The feasible region is empty. Z = 2x + 3y has no maximum value (the LPP has no feasible solution).
Example 2
Question: Maximise \( Z = 3x + 2y \) subject to the constraints \( -x + y \ge 1 \), \( x - 2y \le 2 \), \( x \ge 0 \), \( y \ge 0 \). Show that Z has no finite maximum.
Solution: We rewrite the constraints in standard form and graph them. Constraint 1: \( -x + y \ge 1 \), i.e. \( y \ge x + 1 \). Boundary line: \( y = x + 1 \), passing through \( (0, 1) \) and \( (-1, 0) \). Since \( x \ge 0 \), the relevant intercept with \( x = 0 \) gives the point \( (0, 1) \). The feasible side is above this line. Constraint 2: \( x - 2y \le 2 \), i.e. \( y \ge \frac{x - 2}{2} \). Boundary line: \( x - 2y = 2 \), i.e. \( y = \frac{x-2}{2} \), passing through \( (2, 0) \) and \( (0, -1) \). The feasible side (with \( y \ge 0 \)) is above or on this line. Finding the corner of the feasible region: we find the intersection of \( y = x + 1 \) and \( x - 2y = 2 \). Substituting \( y = x + 1 \) into \( x - 2y = 2 \): \[ x - 2(x + 1) = 2 \] \[ x - 2x - 2 = 2 \] \[ -x = 4 \] So \( x = -4 \). Since \( x \ge 0 \), this intersection point is not in the feasible region. For \( x \ge 0 \): Constraint 1 requires \( y \ge x + 1 \ge 1 \). At \( x = 0 \): \( y \ge 1 \). Constraint 2 at \( x = 0 \): \( 0 - 2y \le 2 \Rightarrow y \ge -1 \), which is always satisfied when \( y \ge 0 \). So the feasible region is the set of all \( (x, y) \) with \( x \ge 0 \), \( y \ge x + 1 \) (and constraint 2 is automatically satisfied for large \( y \)). The feasible region is unbounded (as \( x \to \infty \) with \( y = x + 1 \), the region extends without bound). Checking Z along the boundary \( y = x + 1 \) with \( x \ge 0 \): \[ Z = 3x + 2(x + 1) = 5x + 2 \] As \( x \to \infty \), \( Z \to \infty \). So Z is unbounded above. The lowest point of the feasible region occurs at \( x = 0 \), \( y = 1 \), giving \( Z = 3(0) + 2(1) = 2 \). Therefore, Z has a minimum value of 2 at the point \( (0, 1) \), but Z has no finite maximum value — it can be made as large as desired within the feasible region.
Answer: Z = 3x + 2y has no finite maximum (Z is unbounded above). The minimum value of Z is 2, attained at (0, 1).