PROBLEMS* sing Linear smming To the left of each of the following problems (or their parts), we have inserted a D (for Demo), I ms: The (for Interactive routine), or A (for Automatic routine) whenever a corresponding routine listed ex Method above can be helpful, An asterisk on the I or A indicates that this routine definitely should be used (unless your instructor gives you contrary instructions) and the printout from this routine is all that needs to be turned in to show your work in executing the algorithm. An asterisk on the problem number indicates that at least a partial answer is given in the back of the book. 4.1-1.* Consider the linear programming model (given in the back of the book) that was formulated for Prob. 3.2-3. (@ Use graphical analysis to identify all the corner-point solutions for this model. Label each as either feasible or infeasible. (b) Calculate the value of the objective function for each of the CPF solutions. Use this information to identify an optimal solution. (©) Use the solution concepts of the simplex method given in Sec. 4.1 to identify which sequence of CPF solutions might be examined by the simplex method to reach an ‘optimal solution. (Hint: There are nwo alternate sequences to be identified for this _ particular model.) “\ 41-2) Repeat Prob. 4.1-1 for the following problem. : Maximize Z = x, + 2x2, subject to x +3m <8 xt ms and n20, 20. Repeat Prob.-4.1-1 for the following model. Maximize Z = 3x, + 2x2, x <4 a +3515 2x + m= 10 and x20, 20. 4.1-4, Label each of the following statements about linear programming problems as true or false, and then justify your answer. (@) For minimization problems, if the objective function evaluated at a CPF solution is no larger than its value at every adjacent CPF solution, then that solution is optimal. (®) Only CPF solutions can be optimal, so the number of optimal solutions cannot exceed the number of CPF solutions, _ © IE multiple optimal solutions exist, then an optimal CPF solution may have an adja- i ‘cent CPF solution that also is optimal (the same value of Z), © * Problems 4.3-5, 43-6, 44-6 to 44-8, 4 nermistion fem neevic 46-10, 46-12, 4.7-4, and 4.7-§ haye heen a?aniad seaabe @® Dipindai dengan CamScanner139 tional constraints in augmented 4) obicha BF solution by calculating the values of the slack variables, 1, use the values of the variables to identify the nonbasic variables and the basic variables. {o) For each BF solution, demonstrate (by plugging in the solution) that, after the nonba- sic variables are set equal to zero, this BF solution also is the simultaneous solution of the system of equations obtained in part (a). 42-2. Reconsider the model in Prob, 4,1-2. Follow the instructions of Prob. 4.2-1 for «ats (@), (b), and (c). (d) Repeat part (b) for the comer-point infeasible solutions and the corresponding basic infeasible solutions. (e) Repeat part (c) for the basic infeasible solutions. 4.2.3, Follow the instructions of Prob. 4,2-1 for the model in Prob. 4.1-3. 1A 43-1. Use the simplex method (in algebraic form) to solve the model in Prob. 4.1-1. 43-2, Reconsider the model in Prob. 4.1-2. (a) Apply the simplex method (in algebraic form) by hand to solve this model. 4.) Repeat part (a) with the corresponding interactive routine in your OR Courseware. «© Verify the optimal solution you obtained by using the corresponding automatic rou- tine in your OR Courseware. «43:3, Follow the instructions of Prob, 43-2 for the model in Prob. 4.1-3 wy, Use the simplex method (in algebraic form) to solve the following problem. Maximize Z = 4x, + 3x2 + 6x3, subject 10 3y + x +3x 530 2x, + 2g + 3x5 = 40 and u Ke m=0 20, 20. (oF the’ simplex method (in algebraic form) to solve the following problem. / Maximize Z = x1 + 2x + 4x5, subject to 3x, + ay + Sxy 510 xtdnt ns 8 2x, +2xy5 7 ad — 20, 420, 420. “te 4 43-6. Use the simplex method (in algebraic form) to solve the following problem, a Maximize Z = 41 + 2xz + 2s, ibject to $x) + 2x + 3xy 5 15 ay + day + 2xy 512 2x, + us 8 @ Dipindai dengan CamScannersubject to dy ty+ x20 any +4) + 2a 5 30 and 20, 20, 20. You are given the information that the nonzero variables in the optimal solution are x, and 3. (@) Describe how you can use this information to adapt the simplex method to solve this problem in the minimum possible number of iterations (when you start from the usual initial BF solution). Do not actually perform any iterations. () Use the procedure developed in part (a) to solve this problem by hand. (Do not use x OR Courseware.) —~ (( 43-8. donsider the following problem. — / - Maximize Z= 2x + 4x2 + 3x3, subject to a1 + 3x, + 2x5 = 30 nt nt usu 3a, + 5x + 3x = 60 and 120, 20, 20. You are given the information that x; > 0, x2 = 0, and x3 > 0 in the optimal solution. (@) Describe how you can use this information to adapt the simplex method to solve this problem in the minimum possible number of iterations (when you start from the usual initial BF solution). Do not actually perform any iterations, (b) Use the procedure developed in part (a) to solve this problem by hand. (Do not use your OR Courseware.) 4.3-9. Label each of the following statements as true or false, and then justify your answer by referring to specific statements (with page citations) in the chapter. (a) The simplex method’s rule for choosing the entering basic variable is used because it always leads to the best adjacent BF solution (largest Z). (b) The simplex method's minimum ratio rule for choosing the leaving basic variable is used because making another choice with a larger ratio would yield a basic solution that is not feasible. (©) When the simplex method solves for the next BF solution, elementary algebraic operations are used to eliminate each nonbasic variable from all but one equation (its equation) and to give it a coefficient of +1 in that one equation. D,I*,A 44-1, Repeat Prob. 4.3-1, using the tabular form of the simplex method. D, 1", a* 44-2, Repeat Prob. 4.3-2, using the tabular form of the simplex method. D,1*, A* 4.4-3. Repeat Prob. 4.3-3, using the tabular form of the simplex method, @® Dipindai dengan CamScanner= ‘onsider the following problem. 141 Maximize = Z = 2x, +, 4/ Problems ates 40 4x, +x, 5 100 J 20, 20. (a) Solve this problem graphically by hand. Also identify all the CPF solutions. + @®) Now use your OR Courseware to solve this problem graphically. (©) Use hand calculations to solve this problem by the simplex method in algebraic form. 4 (d) Now use your OR Courseware to solve this problem interactively by the simplex method in algebraic form. (©) Use hand calculations to solve this problem by the simplex method in tabular form. 4 Cf) Now use your OR Courseware to solve this problem interactively by the simplex 10d in tabular form. ( 4.4-5.)Repeat Prob. 44-4 for the following problem. wR Maximize Z = 2x, + 3x, west 10 a+ 2m 530 at 520 "7 + 444-6. Consider the following problem. 120, 20. Maximize Z = 2x, + 4xz + 32, jest t0 3x + 4x2 + 2x5 = 60 2y + x + 2x) 540 ay + 3xz + 2x5 = 80 20, 20, 420. (a) Solve by the simplex method in algebraic form. fi Ive by the simplex method in tabular form. 44-7, \ sider the following problem. Maximize Z= 3x, + Sap + 6x, feet to It mt usd ay +2 + xy 54 nt ntiysd ix tax tax <3 @ Dipindai dengan CamScanner142 4/ Solving Linear Programming Problems: The Simplex Method M20 20, 420. (a) Solve by the simplex method in algebraic form. 44 )-Salve by the simplex method in tabular form D.1%.4 AA, Consider the following problem. a Maximize = Z= 2x, — x +43, subject to nora tins 4 hte 510 oR Hs 7 and M20, 20, 20. (a) Solve by the simplex method in algebraic form. «- @)-Solve by the simplex method in tabular form. D.1*,4 4,49, Use the simplex method (in tabular form) to solve the following problem.- Maximize = Z= 2x — x +43, subject to By tat 456 ny- x ty Sd ntn- 452 420 B=0, 420. the simplex method to solve the following problem, Maximize = Z= —x, +x, + 2xs, subject to xt2m- 45520 Oxy + 4 + 2x5 = 60 2x, + 3x + x9 550 and 120, 20, H2=0, 45-1, Consider the following statements about linear programming and the simplex ‘method. Label each statement as true or false, and then justify your answer. (@) Ina particular iteration of the simplex method, if there is a tie for which variable should be the leaving basic variable, then the next BF solution must have at least one basic variable equal to zero. (0) If there is no leaving basic variable at some iteration, then the problem has no feasible solutions. (C) If at least one of the basic variables has a coefficient of zero in row O of the final tableau, then the problem has multiple optimal solutions. @ Dipindai dengan CamScanner(@) Ifthe problem has multiple optimal solutions, then the problem must have a bounded sete region, 143 4/ Problems 44 45-% Consider the following problem. u Maximize Z= Sry +. + 3x + dtu opto } Xy — Day + day + 3x45 20 4x) + 6x2 + Say — dry 5 40 2x, — Bry + 3xy + 8x4 5 50 N20 20 20, 420, ie the simplex method to demonstrate that Z is unbounded. 4.5-3. A basic property of any linear programming problem with a bounded feasible «gion is that every feasible solution can be expressed as a convex combination of the CPF «tions (perhaps in more than one way). Similarly, for the augmented form of the problem, ety feasible solution can be expressed as a convex combination of the BF solutions. (a)_ Show that any convex combination of any set of feasible solutions must be a feasible solution (so that any convex combination of CPF solutions must be feasible), (6) Use the result quoted in part (a) to show that any convex combination of BF solutions must be a feasible solution, 45-4. Using the facts given in Prob. 4.5-3, show that the following statements must be ne for any linear programming problem that has a bounded feasible region and multiple optimal ‘solutions: (a) Every convex combination of the optimal BF solutions must be optimal. “} ©) No other feasible solution can be optimal. 1h 458° Chnsider the following problem. Maximize tm s3 xytys2 420, forj=1,2,3,4 %€ the simplex method to find all the optimal BF solutions. i | 4.6-1.* Consider the following problem. = Maximize Z = 2x, + 3xn “Yject to ay t2ns4 | mt m a3 420, 20, (a) Solve this problem graphically. @® Dipindai dengan CamScanner144 4/Solving Linear ‘Programming Problems: The ‘Simplex Method (®) Using the Big M method, construct the complete first simplex tableau for the simplex a and identify the corresponding initial (artificial) BF solution. Also identify I entering basic variable and the leaving basic variable. BA (oh sane by the simplex method, 46-2, Consider the following problem. Maximize Z = 4xy + 2xy + 3x5 + Sta subject to Duy + Bay + 4xy + Dey = 300 Bx + oxy t xy + Sty = 300 and 20, forj= 12,34. (a) Using the Big M method, construct the complete first simplex tableau for the simplex method and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable, 1,A (0) Solve by the simplex method. (©) Using the two-phase method, construct the complete first simplex tableau for phase 1 and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable. 1d) Perform phase 1. (©) Construct the complete first simplex tableau for phase 2. 1,4 (f) Perform phase 2, (g) Compare the sequence of BF solutions obtained in part (b) with that in parts (d) and (f). Which of these solutions are feasible only for the artificial problem obtained by lucing artificial variables and which are actually feasible for the real problem? 4,6-3. Consider the following problem. } j Minimize Z = 3x, + 2x, subject to 2x + 210 -3n +2n< 6 nt me 6 and 120, 220, 1 (a) Solve this problem graphically. (®) Using the Big M method, construct the complete first simplex tableau for the simplex method and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable. 1A (c)_Solve by the simplex method. «4 Consider the following problem. \ Minimize Z= 2x, + 3x2 +23, ‘Subject to xy tar +228 3x, + 2x 26 @® Dipindai dengan CamScanner420, m20, 20. ) Reformulate this problem to fit our standard form for a linear programming model presented in Sec. 3.2. .) Using the Big M method, solve by the simplex method. “> Using the two-phase method, solve by the simplex method. * }) Compare the sequence of BF solutions obtained in parts (b) and (c). Which of these solutions are feasible only for the antificial problem obtained by introducing artificial variables and which are actually feasible for the real problem? 4o5- For the Big M method, explain why the simplex method never would choose an |Jvangble to be an entering basic variable once all the artificial variables are nonbasic. 46-6,/Consider the following problem, Maximize Z = 2x, + 5x) + 345, x17 2m +45 220 2xy + Ary +25 = 50 120, M20, 4520, (a) Using the Big M method, construct the complete first simplex tableau for the simplex method and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable. (6) Solve by the simplex method. (6) Using the two-phase method, construct the complete first simplex tableau for phase 1 and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable. (d) Perform phase 1. (¢) Construct the complete first simplex tableau for phase 2. (f) Perform phase 2. (g) Compare the sequence of BF solutions obtained in part (b) with that in parts (4) and ). Which of these solutions are feasible only for the artificial problem obtained by introducing artificial variables and which are actually feasible for the real problem? 46-7. Consider the following problem. . Minimize Z= 2x + xp + 3x5, Pett Say + Oxy + xy = 420 3ny + Day + 5x3 = 280 120, 220, 420. ° the two-phase method, solve by the simplex method. 46-8%. Consider the following problem. Minimize Z = 3x; + 2x2 + 4xy. et to dat mt3xy= 60 145 4/ Problems @® Dipindai dengan CamScannerSolving Linear gramming blems: The plex Method Buy + Buy + Sxy = 120 and 20, 20, m0. 1A (a) Using the Big M method, solve by the simplex method. 1.4 (b) Using the two-phase method, solve by the simplex method. (c) Compare the sequence of BF solutions obtained in parts (a) and (b). Which of these lutions are feasible only for the artificial problem obtained by introducing artificial ables and which are actually feasible for the real problem? 4.6.9. Gonsider the following problem. Minimize Z = 3x) + 2p + Tx, subject to stm = 10 2xy — x +45 = 10 and 120, 420, 120. 1A (a) Using the Big M method, solve by the simplex method. 1+, 4 (b) Using the two-phase method, solve by the simplex method. (©) Compare the sequence of BF solutions obtained in parts (a) and (b). Which of these __-solutions are feasible only for the artificial problem obtained by introducing artificial g les and which are actually feasible for the real problem? 4.6-10. Consider the following problem, Minimize Z = 3x, + 2x. + x3, mtn =7 3x, + +22 10 1120, 20, 520, 1*,A (a) Using the Big M method, solve by the simplex method. 1*,A (b) Using the two-phase method, solve by the simplex method. (c) Compare the sequence of BF solutions obtained in parts (a) and (b), Which of these solutions are feasible only for the artificial problem obtained by introducing artificial variables and which are actually feasible for the real problem? 4.6-11, Label each of the following statements as true or false, and then justify your answer, (@) When a linear programming model has an equality constraint, an artificial variable is introduced into this constraint in order to start the simplex method with an obvious initial basic solution that is feasible for the original model. (6) When an artificial problem is created by introducing artificial variables and using the Big M method, if all artificial variables in an optimal solution for the artificial prob- lem are equal to zero, then the real problem has no feasible solutions. (©) The two-phase method is commonly used in practice because it usually requires fewer iterations to reach an optimal solution than the Big M method does. 4.6-12. Consider the following problem. @® Dipindai dengan CamScannerMaximize xy t dey + ry, dy om t us 5 mn - m+ 2y 510 soegtivity constraint for xy). «) Reformulate this problem so all variables have nonnegativity constraints, 8) Salve'by the simplex method. 4613" Consider the following problem. Maximize Z= -x + 4%, n+ ns 6 ntins 4 n2-3 wer bound constraint for x1). (0) Solve this problem graphically, (0) Reformulate this problem so that it has only two functional constraints and all vari- ables have nonnegativity constraints. 40) Solve by the simplex method. 4614, Consider the following problem. Maximize Z= -x, +2x) +2, 0 3x + x.5120 nonn 4s 80 3x +2 + 2x5 < 100 ‘wnnegativity constraints). (2) Reformulate this problem so that all variables have nonnegativity constraints. 4 (b) Solve by the simplex method. 46-15. This chapter has described the simplex method as applied to linear programming ‘tems where the objective function is to be maximized. Section 4.6 then described how to "erta minimization problem to an equivalent maximization problem for applying the simplex “tod, Another option with minimization problems is to make a few modifications in the “actions for the simplex method given in the chapter in order to apply the algorithm directly. (4) Describe what these modifications would need to be. (b) Using the Big M method, apply the modified algorithm developed in part (a) to solve the following problem directly by hand. (Do not use your OR Courseware.) Minimize Z = 3x, + 8xy + Ss, et to 3x + 4x = 70 34 + Sq + 2xy = 70 @® Dipindai dengan CamScanner 147 4/ Problems148 4/Sobing Linear Problems: The Simplex: Method and H20, 20, 20. 4.6-16. Consider the following problem. Maximize Z = —2xy +42 ~ 445 + 32a, subject to nt mt3y tus 4 * - yt uel 2ay + = 2 ayt2nt at 2xy= 2 and 20, 20, 420 (no nonnegativity constraint for 2). (@) Reformulate this problem to fit our standard form for a linear programming model presented in Sec. 3.2. (b) Using the Big M method, construct the complete first simplex tableau for the simplex method and identify the corresponding initial (artificial) BF solution. Also identify the initial entering basic variable and the leaving basic variable. (©) Using the two-phase method, construct row 0 of the first simplex tableau for phase 1. A+ d)~Use the automatic routine for the simplex method in your OR Courseware to solve is problem. mA 46-17.) Consider the following problem. - Maximize Z= 4x, + Sx. + 3x3. subject to a+ ay + 2x) = 20 15x + 6x9 — 5x5 = 50 4 + 3x + Sx = 30 H20 m=0, m=O. Use the simplex method to demonstrate that this problem does not possess any feasible solutions. 4.7-1, Refer to the allowable range to stay feasible given in Fig. 4.10 for the respective right-hand sides of the Wyndor Glass Co. problem given in Sec. 3.1. Use graphical analysis to demonstrate that each given allowable range is correct, 4.7-2, Reconsider the model in Prob. 4.1-2. Interpret the right-hand side of the respective functional constraints as the amount available of the respective resources. 1 (q) Use graphical analysis as in Fig, 4.8 to determine the shadow prices for the respective resources. 1 (®) Use graphical analysis to perform sensitivity analysis on this model. In particular, check each parameter of the model to determine whether it is a sensitive parameter (a Parameter whose value cannot be changed without changing the optimal solution) by examining the graph that identifies the optimal solution, (©) Use graphical analysis as in Fig. 4.9 to determine the allowable range for each ¢ value (coefficient of x, in the objective function) over which the current optimal solution will remain optimal. @® Dipindai dengan CamScanner(¢) Changing just one b, value (the right-hand side of functional constraint /) wil shift the 149 corresponding constraint boundary, If the current optimal CPF solution ies on this constraint boundary, this CPF solution also will shift. Use graphical analysis to deter- 4/Prebhens mine the allowable range for each b value over which this CPF solution will remain feasible, \@) Verify your answers in parts (a), (c), and (d) by running the automatic routine for the simplex method in your OR Courseware and identifying the section in the output that corresponds to each of these parts, 47-3, Repeat Prob. 4.7-2 for the model in Prob. 4.13 47-4, You are given the following linear programming problem. Maximize Z = 4x, + 2r, a0 2x S16 (resource 1) xy 43x17 (resourve 2) 2S 5 (resource 3) 20, 20. (a) Solve this problem graphically. (&) Graphically find the shadow prices for the resources. (©), Determine how many additional units of resource 1 would be needed to increase the ‘optimal value of Z by 15. 47-5. Consider the following problem. Maximize Z= x, — 7x; + 3x3, wtto 2+ m—m<4 — (fesource 1) 4x—3x. $2 (resource 2) 3x +2 +4553 (resource 3) 420, 420, 420. (a) Solve by the simplex method. (6) Identify the shadow prices for the three resources and describe their significance. 4.7-6.* Consider the foitowing problem. Maximize Z= 2x; - 2x + 34s, witto =x tyt 45 4 (resource 1) 2xy- a+ x)= 2 (resource 2) ay tx +3x512 (resource 3) 120, 420 420. @ Dipindai dengan CamScanner