Abstract
This paper compares the performances of genetic algorithm with various classical algorithms in solving fractional programming. Genetic algorithm is one of the new forms of algorithms for solving optimization problems, which may not be efficient but a generic way to solve nonlinear optimization problems. The traditional optimization algorithms have difficulty in computing the derivatives and second order partial derivatives, i.e., Hessian for the fractional function. The issues of discontinuity seriously affects traditional algorithm. There are large numbers of classical methods for searching the optimum point of nonlinear functions. The classical search algorithms may be largely classified as gradient based methods and nongradient methods. Here, a comparative performance analysis of different algorithms is made through a newly defined function called algorithmic index. An algorithm based on heuristics for computation of gewicht vector required to derive algorithmic index has also been proposed here.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Charnes, A.C.W.: An explicit general solution in linear fractional programming. Naval Res. Logist. Quart. 20, 449–467 (1973)
Bitran, G.R., Novaes, A.G.: Linear programming with fractional objective function. Oper. Res. 21, 22–29 (1973)
Birtan, G.R., Magnanti, T.L.: Duality and sensitivity analysis of fractional objective function. Oper. Res. 24, 675–699 (1976)
Swarup, K.: Linear fractional programming. Oper. Res. 13(6), 1029–1036 (1965)
Dantzig, G.B.: Linear Programming under uncertainty. Manage. Sci. 1(3 and 4), 197–206 (1955)
Dantzig, G.B., Mandansky, A.: on solution of two stage linear programming under uncertainty. Barkley Symp. Maths Stat. 1(3 and 4), 165–176 (1961)
Bazalinov, E.B.: Linear Fractional Programming. Kluwer Academic Publishers, Dordrecht (2003)
Bajalinov, E.: Scaling problems in linear fractional programming. In: proceeding of 10th international conference on operation research, vol. 3, no. 1, pp. 22–24 (2004)
Shohrab, E., Morteza, P.: Solving the Interval Valued Fractional Programming. Am. J. Comput. Math. 2(1), 51–55 (2012)
Barros, A.I., Frenk, J.B., Schaible, S., Zhang, S.: A new algorithm for generalised fractional programming. Math. Program. 72(2), 147–175 (1996)
Gogia, N.: Revised simplex algorithm for linear fractional programming problem. Math. Student 36(1), 55–57 (1969)
Horvath, I.: AsupraprogramliriifracJionare lineare cu restricJii suplimentare. Informatica pentru Conducere, pp. 101–102 (1981)
Stahl, J.: Two new methods for solution of hyperbolic programming. In: Publications of the Mathematical Institute of Hungarian Science, vol. 9, no. B, pp. 743–754 (1964)
Stancu-Minasian, I.M.: Stochastic Programming with MultiObjective Function. D. Reidel Publishing Company, Dordrecht (1984)
Buhler, W.: A note on fractional interval programming. Oper. Res. A-B 19, 1, 29–36 (1975); Z(19), 29–35 (1975)
Robins, H., Monro, S.: A stochastic approximation method. Ann. Math. Stat. 22, 400–407 (1951)
Costa, A., Jones, O., Kroese, D.: Convergence properties of the cross entropy method for discrete optimization. Oper. Res. Lett. 35, 573–580 (2007)
Zhang, Q., Muhlenbein, H.: On the convergence of a class of estimation of distribution algorithm. IEEE Trans. Evol. Comput. 8, 127–134 (2004)
Binglsley, P.: Convergence of probability measures. John Wiley and Sons, New York (1999)
Box, G.E.: Evolutionary operation: a method for increasing industrial productivity. Appl Stat 6, 81–101 (1957)
Brent, R.P.: Algorithms for Minimization without derivatives. Printice Hall, EngleWoods Cliffs (2002)
Mifflin, R., Strodiot, J.J.: A Bracketing technique to ensure desirable convergence in univariate minimisation. Math. Prog. 17, 100–117 (1975)
Mifflin, R., Strodiot, J.J.: A rapidly convergent five-point algorithm for univariate minimisation. Math. Prog. 62, 299–319 (1993)
Box, G.P., Wilson, K.B.: On the experimental attainment of optimal conditions. Stat. Soc. 13, 1–13 (1951)
Box, G.E.P., Draper, N.R.: Evolutionary Operation: A Statistical Method For Process Improvement. Wiley, New York (1998)
Hooke, R., Jeeves, T.A.: Direct search solution for numerical and statistical problem. ACM 8, 212–219 (1961)
Nelder, J.A., Mead, R.: A simplex method for function minimisation. Comput. J. 7, 308–313 (1965)
Fletcher, R.: Practical Methods for Optimisation. John Wiley and Sons, Chichester (1987)
Murray, W., Wright, M.H., Gill, P.E.: Practical Optimization. Academic Press, London (1981)
Gabay, D.: Reduced Quasi Newton method with feasibilty improvement for nonlinear constrained optimisation. Math. Prog. Stud. 16, 18–44 (1982)
Fletcher, R.: Conjugate Gradient Methods for Indefinite Systems. Numer. Anal. Rep. 5, 11 (1975)
Zangwill, W.: Nonlinear Programming: A Unified Approach. Printice Hall, Englewood Cliffs (1969)
Holland, J.H.: Hierarchical description of universal spaces and adaptive systems. Tech. Rep ORA Project 01252 (1968)
Goldberg, D.: Genetic Algorithms in Search, Optimization, and Machine Learning. Addison-Wesley, Reading (1989)
Grefenstette, J.J.: Deception considered harmful. In: Whitley, L.D. (ed.) Foundations of genetic algorithms (1993)
Fogel, D.G.: Schema processing under proportional selection in the presence of random effects. IEEE Trans. Evol. Comput. 1(4), 290–293 (1997)
Radcliffe, N.J.: Schema Processing. In: Handbook of evolutionary computation, pp. B2.5–1.10. Oxford University Press (1997)
Acknowledgements
Special thanks go to my guides Dr. Sujyasikha Das and Dr. Swarup Prasad Ghosh for inspiring me to write the paper and implementing the scenario in Matlab. The paper would remain unfinished if I don’t convey my regards and heartfelt thanks to Dr. Nabendu Chaki for relentless support to my academics. He has been the driving force for all the activities.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
Algorithm for Random Search
-
Step 1:
Choose initial x0, z0, ϵ such that the minimum lies in (x0 − 1/2z0, x0 + 1/2z0). For each Q block, set q = 1 and p = 1.
-
Step 2:
For i = 1,2…N, create points using uniform distribution of m in the range (–0.5,0.5). Set x (p)i  = x q−1i  + mz q−1i .
-
Step 3:
If x(p) is infeasible and p < P, repeat Step 2. If x(p) is feasible, save x(p) and f(x(p)). Increment p and repeat step 2;
Else if p = P, set xq to be the point that has lowest f(x(p)) over all feasible x(p) including xq−1
And reset p = 1.
-
Step 4:
Reduce the range via z qi  = ϵ z q−1i .
-
Step 5:
If q > Q. Stop.
Else increment q and continue to Step 2.
Box’s Evolutionary Algorithm
-
Step 1:
Choose initial point. Choose size reduction step δi and termination criteria ϵ ..
-
Step 2:
If δi < ϵ . STOP.
-
Step 3:
Else create 2N points by adding and subtracting δi from each variable at the initial point.
-
Step 4:
Compute function values at all 2N points. Find the optimum among these points. Set it as initial point for next iteration.
-
Step 5:
Reduce size of the step to δi/2 and go to Step 2.
Hooke’s Jeeves’ Algorithm
-
Step 1:
Initial point is selected and objective function is evaluated.
-
Step 2:
Search is made in the direction of each dimension by a step size Si to find lowest of functional value.
-
Step 3:
In case the function value does not decrease in any direction, the step size is reduced and fresh search is made.
-
Step 4:
If the value of objective function reduces, a new initial point is found as follows:
X (k+1)i,o  = X k+1i + θ (X k+1i  − X ki ), θ  > 1.
-
Step 5:
This search continues till the termination criteria is met, i.e., θ  < ϵ .
Gradient Descent Method
-
Step 1:
Choose initial point x(0) and termination parameters ϵ 1 and ϵ 2.
-
Step 2:
Compute first derivative \(\nabla {\text{f}}\left( {x^{k} } \right).\)
-
Step 3:
If \(\left| {\left| {\nabla {\text{f}}\left( {x^{k} } \right)} \right|} \right| \le\) ϵ 1 STOP.
Else go to next step
-
Step 4:
By unidirectional search, find α k such that f(x(k+1)) = f(xk − α k \(\nabla {\text{f}}\left( {x^{k} } \right)\)) is minimum. One criteria for termination is | \(\nabla {\text{f}}\left( {x^{k + 1} } \right).\nabla {\text{f}}\left( {x^{k} } \right)\) | ≤  ϵ 2.
-
Step 5:
If \(\frac{{||x^{k + 1} - x^{k} ||}}{{\left| {\left| {x^{k}} \right|} \right|}} \le\) ϵ 1, then STOP.
Else set k = k + 1, go to step 2.
Genetic Algorithm
-
Start and generate a random population of size n.
-
Fitness: Evaluate fitness of each chromosome.
-
New Population: Create new population by repeating the steps below
-
Select two parent chromosomes from the population according to best fitness.
-
Cross over the parents, with a crossover probability to form new population.
-
With a mutation probability, mutate the new offspring.
-
Add the new offspring in the population.
-
Replace: Use the new generation for next iteration.
-
Test: Check termination criteria.
-
Loop: Go to step 2.
MATLAB Code for Estimating Gewicht Vector
Rights and permissions
Copyright information
© 2016 Springer India
About this chapter
Cite this chapter
Roy, D., Das, S.S., Ghosh, S. (2016). Comparative Analysis of Genetic Algorithm and Classical Algorithms in Fractional Programming. In: Chaki, R., Cortesi, A., Saeed, K., Chaki, N. (eds) Advanced Computing and Systems for Security. Advances in Intelligent Systems and Computing, vol 396. Springer, New Delhi. https://doi.org/10.1007/978-81-322-2653-6_17
Download citation
DOI: https://doi.org/10.1007/978-81-322-2653-6_17
Published:
Publisher Name: Springer, New Delhi
Print ISBN: 978-81-322-2651-2
Online ISBN: 978-81-322-2653-6
eBook Packages: EngineeringEngineering (R0)