import numpy as np import matplotlib.pyplot as plt from scipy.optimize import linprog

Define the constraints

x = np.linspace(0, 5, 400) y1 = (30 - 3x) / 6 # Constraint: 3x + 6y <= 30 y2 = (12 - 2x) / 2 # Constraint: 2x + 2y <= 12

Boundaries of x and y

x_bounds = [0, 4] y_bounds = [0, max(np.minimum(y1, y2))]

Intersection points (vertices of feasible region)

points = [ (0, 0), # Origin (4, 0), # x-boundary (0, 5), # y-boundary of 3x + 6y (2, 3), # Intersection of 3x + 6y = 30 and 2x + 2y = 12 ]

Function to maximize

c = [-5, -3] # Maximize 5x + 3y A = [[3, 6], [2, 2], [1, 0], [0, 1]] b = [30, 12, 4, 0]

Solve linear programming problem

res = linprog(c, A_ub=A, b_ub=b, bounds=(x_bounds, y_bounds), method='highs')

Plotting the feasible region

plt.figure(figsize=(10, 8)) plt.plot(x, y1, label="3x + 6y = 30", color="blue") plt.plot(x, y2, label="2x + 2y = 12", color="green") plt.fill_between(x, np.maximum(0, np.minimum(y1, y2)), alpha=0.3, color='grey', label='Feasible Region')

Plot the vertices of the feasible region

vertices = np.array(points) plt.scatter(vertices[:, 0], vertices[:, 1], color="red", zorder=5, label="Vertices")

Highlight optimal solution

plt.scatter(res.x[0], res.x[1], color="purple", zorder=10, label="Optimal Solution")

Labels and legend

plt.xlim(-1, 5) plt.ylim(-1, 6) plt.xlabel("x") plt.ylabel("y") plt.axhline(0, color='black',linewidth=0.5) plt.axvline(0, color='black',linewidth=0.5) plt.grid(color = 'gray', linestyle = '--', linewidth = 0.5) plt.title("Feasible Region and Optimal Solution") plt.legend() plt.show()

Display the result

res.x, res.fun <<ImageDisplayed>>(array([ 4., -0.]), -20.0)(array([ 4., -0.]), -20.0)successThe optimal solution to the linear programming problem is:

• Optimal Point: $x = 4, y = 0$
• Maximum Value of $P$: $P = 5x + 3y = 20$

Explanation:

• The graph above illustrates the feasible region formed by the constraints.
• The vertices (corners) of the feasible region were evaluated, and the point $(4, 0)$ yielded the maximum value of $P$.
• The constraints and boundaries were respected in defining the region and solving the problem.

Let me know if you need further clarification or additional calculations!

---

Related Questions:

1. What happens if we change the coefficients in the objective function?
2. How can we solve a similar problem using alternative optimization techniques?
3. What are the implications of having unbounded regions in linear programming?
4. How would the solution differ if one of the constraints is removed?
5. Can we apply these methods to problems with more than two variables?

Tip:

Always check the feasibility of the solution by substituting the values of $x$ and $y$ back into the constraints.