1 / 30

Sensitivity Analysis

Sensitivity Analysis. How will a change in a coefficient of the objective function affect the optimal solution? How will a change in the right-hand side value for a constraint affect the optimal solution?. Pet Food Co. – Linear Equations. Pet Food Co. – Graph Solution. Line 3. Line 2.

lin
Télécharger la présentation

Sensitivity Analysis

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Sensitivity Analysis • How will a change in a coefficient of the objective function affect the optimal solution? • How will a change in the right-hand side value for a constraint affect the optimal solution? Linear Programming

  2. Pet Food Co. – Linear Equations Linear Programming

  3. Pet Food Co. – Graph Solution Line 3 Line 2 Linear Programming

  4. Pet Food Co. – Optimal Solution • Extreme Point is optimal if: • Slope of Line 3 <= Slope of objective function <= Slope of Line 2 Linear Programming

  5. Pet Food Co. – Calculate Slope of Line 3 1P1 + 1P2 >= 500 1P2 >= -1P1 + 500 P2 >= -P1 + 500 Slope of Intercept of Line 3 Line 3 on P2 axis Linear Programming

  6. Pet Food Co. – Calculate Slope of Line 2 0P1 + 1P2 >= 200 1P2 >= -0P1 + 200 Slope of Intercept of Line 2 Line 2 on P2 axis Linear Programming

  7. Pet Food Co. – Optimal Solution • Extreme Point 4 is optimal if: • -1 <= Slope of objective function <= 0 Linear Programming

  8. Calculating Slope-Intercept • General form of objective function • Z = CP1P1 + CP2P2 • Slope-intercept for objective function • P2 = -(CP1/CP2) P1 + Z/CP2 Slope of Intercept of Obj. Function Obj. Function on x2 axis Linear Programming

  9. Pet Food Co. – Optimal Solution • Extreme Point is optimal if: • -1 <= -(CP1/CP2) <= 0 Or • 0 <= (CP1/CP2) <= 1 Linear Programming

  10. Pet Food Co. – Compute the Range of Optimality • Extreme Point is optimal if: • 0 <= (CP1/CP2) <= 1 • Compute range for CP1, hold CP2constant • 0 <= (CP1/8) <= 1 Linear Programming

  11. Pet Food Co. – Compute the Range of Optimality • From the left-hand inequality, we have • 0 <= (CP1/8) • Thus, • 0 <= CP1 Linear Programming

  12. Pet Food Co. – Compute the Range of Optimality • From the right-hand inequality, we have • (CP1/8) <= 1 • Thus, • CP1 <= 8 Linear Programming

  13. Pet Food Co. – Compute the Range of Optimality • Summarizing these limits • 0 <= CP1 <= 8 Linear Programming

  14. Pet Food Co. – Compute the Range of Optimality • Extreme Point is optimal if: • 0 <= (CP1/CP2) <= 1 • Compute range for CP2, hold CP1constant • 0 <= (5/CP2) <= 1 Linear Programming

  15. Pet Food Co. – Compute the Range of Optimality • From the left-hand inequality, we have • 0 <= (5/CP2) • Thus, • (1/5) * 0 <= (1/CP2) • Invert • 5/0 >= CP2 • Division by zero is undefined (infinite). • This means the cost of P2 can increase to infinity without changing the optimal solution Linear Programming

  16. Pet Food Co. – Compute the Range of Optimality • From the right-hand inequality, we have • (5/CP2) <= 1 • Thus, • CP2 >= 5 Linear Programming

  17. Pet Food Co. – Compute the Range of Optimality • Summarizing these limits • 0 <= CP1 <= 8 • 5 <= CP2 <= Infinite Linear Programming

  18. Sensitivity Analysis • How will a change in a coefficient of the objective function affect the optimal solution? • How will a change in the right-hand side value for a constraint affect the optimal solution? Linear Programming

  19. Pet Food Company – Graph Solution Line 3 Line 2 Linear Programming

  20. Pet Food Co. – Range of Feasibility • Constraint 1 – is not binding • Therefore, the shadow price is zero • Slack is 100 • Range of Feasibility • 300 <= Constraint 1 RHS <= Infinite Linear Programming

  21. Pet Food Co. – Change in the Right-hand Side • Constraint 2 – add 1 to right-hand side • 0P1 + 1P2 >= 201 • 1P1 + 1P2 >= 500 • Solve for P1 • -1(0P1 + 1P2 = 201) • 1P1 + 1P2 = 500 • P1 = 299 • Solve for P2 • 1(299) + 1P2 >= 500 • P2 =201 Linear Programming

  22. Pet Food Co. – Change in the Right-hand Side • Solve objective function • z = 5(299) + 8(201) • z = 3103 • Shadow Price • 3103 – 3100 = 3 • Thus cost increases at $3.00 per lb. added of P2 per batch • Conversely, if we decrease lbs. of P2 per batch by 1 the objective function will decrease by $3.00 Linear Programming

  23. Pet Food Co. – Range of Feasibility • Constraint 2 RHS = 200 • Allowable Increase = 300 • Allowable Decrease = 100 • Range of Feasibility • 100 <= Constraint 2 RHS <= 500 Linear Programming

  24. Pet Food Co. – Change in the Right-hand Side • Constraint 3 – add 1 to right-hand side • 0P1 + 1P2 >= 200 • 1P1 + 1P2 >= 501 • Solve for P1 • -1(0P1 + 1P2 = 200) • 1P1 + 1P2 = 501 • P1 = 301 • Solve for P2 • 1(301) + 1P2 >= 501 • P2 = 200 Linear Programming

  25. Pet Food Co. – Change in the Right-hand Side • Solve objective function • z = 5(301) + 8(199) • z = 3097 • Shadow Price • 3105 – 3100 = 5 • Thus cost increases at $5.00 per lb. added of P2 per batch • Conversely, if we decrease lbs. of P2 per batch by 1 the objective function will decrease by $5.00 Linear Programming

  26. Pet Food Co. – Range of Feasibility • Constraint 3 RHS = 500 • Allowable Increase = 100 • Allowable Decrease = 300 • Range of Feasibility • 200 <= Constraint 3 RHS <= 600 Linear Programming

  27. Pet Food Co. – Linear Equations Slack/ Surplus Variables Min 5P1 + 8P2 + 0S1 + 0S2 + 0S3 s.t. 1P1 + 1S1 = 400 1P2 - 1S2 = 200 1P1 + 1P2 - 1S3 = 500 P1, P2, S1 ,S2 ,S3 >= 0 Linear Programming

  28. Pet Food Co. – Slack Variables • For each ≤ constraint the difference between the RHS and LHS (RHS-LHS). It is the amount of resource left over. • Constraint 1; S1 = 100 lbs. Linear Programming

  29. Pet Food Co. – Surplus Variables • For each ≥ constraint the difference between the LHS and RHS (LHS-RHS). It is the amount bt which a minimum requirement is exceeded. • Constraint 2; S2 = 0 lbs. • Constraint 3; S3 = 0 lbs. Linear Programming

  30. Pet Food Co. – Constraint Limits • Range of Feasibility • 300 <= Constraint 1 <= Infinite • 100 <= Constraint 2 <= 500 • 200 <= Constraint 3 <= 600 Linear Programming

More Related