Application of robust optimization for a product portfolio problem using an invasive weed optimization algorithm

Alireza Goli; Hasan Khademi Zare; Reza Tavakkoli-Moghaddam; Ahmad Sadeghieh

doi:10.3934/naco.2019014

Article Contents

2019, Volume 9, Issue 2: 187-209. Doi: 10.3934/naco.2019014

This issue Previous Article Iterative methods for solving large sparse Lyapunov equations and application to model reduction of index 1 differential-algebraic-equations Next Article Homotopy method for matrix rank minimization based on the matrix hard thresholding method

Application of robust optimization for a product portfolio problem using an invasive weed optimization algorithm

1.
Department of Industrial Engineering, Yazd University, Saffayieh, Yazd, Iran
2.
School of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran

Received: February 2018

Revised: August 2018

Published: January 2019

Abstract Full Text(HTML) Figure(8) / Table(11) Related Papers Cited by

Abstract

Product portfolio optimization (PPO) is a strategic decision for many organizations. There are several technical methods for facilitating this decision. According to the reviewed studies, the implementation of the robust optimization approach and the invasive weed optimization (IWO) algorithm is the research gap in this field. The contribution of this paper is the development of the PPO problem with the help of the robust optimization approach and the multi-objective IWO algorithm. Considering the profit margin uncertainty in real-world investment decisions, the robust optimization approach is used to address this issue. To illustrate the real-world applicability of the model, it is implemented for dairy products of Pegah Golpayegan Company in Iran. The numerical results obtained from the IWO algorithm demonstrate the effectiveness of the proposed algorithm in tracing out the efficiency frontier of the product portfolio. The average risk of efficient frontier solutions in the deterministic model is about 0.4 and for the robust counterpart formulation is at least 0.5 per product. The efficient frontier solutions obtained from robust counterpart formulation demonstrate a more realistic risk level than the deterministic model. The comparisons between CPLEX, IWO and genetic algorithm (GA) shows that the performance of the IWO algorithm is much better than the older algorithms and can be considered as an alternative to algorithms, such as GA in product portfolio optimization problems.

Keywords:

Mathematics Subject Classification: Primary: 58F15, 58F17; Secondary: 53C35.

Citation:

Full Text(HTML)

Figure 1. Pseudocode of IWO algorithm

Download: Full-size image PowerPoint slide

Figure 2. An example of solution representation

Download: Full-size image PowerPoint slide

Figure 3. Efficiency frontier of portfolio selection problem with 10 products

Download: Full-size image PowerPoint slide

Figure 4. Efficiency frontier of portfolio selection problem with 50 products

Download: Full-size image PowerPoint slide

Figure 5. Comparison of objective function values obtained from the deterministic model and the robust models with different uncertainty levels for K = 10

Download: Full-size image PowerPoint slide

Figure 6. Comparison of objective function values obtained from the deterministic model and the robust models with different uncertainty levels for K = 50

Download: Full-size image PowerPoint slide

Figure 7. The execution time of CPLEX, IWO and Ga

Download: Full-size image PowerPoint slide

Figure 8. GA and IWO portfolio efficient frontier

Download: Full-size image PowerPoint slide

Table 1. Optimal setting of IWO Algorithm parameters

Parameter	Symbol	Optimal Value
Max iteration	Maxit	100
Number of first population	Npop0	50
Maximum population size	Pmax	100
Minimum seed	Smin	20
Maximum seed	Smax	50
Reducing power of standard deviation (pu)	N	0.01
Initial standard deviation	Sigma_final	0.5
Final standard deviation	Sigma_final	0.1

| Show Table

DownLoad: CSV

Table 2. Computational results of CPLEX, GA and IWO in the small-scale problems

Problem	K	CPLEX		GA			IWO
Problem	K	Z	T	Z	T	GAP(%)	Z	T	GAP(%)
P1	2	-0.678	1.6	-0.678	0.03	0	-0.678	0.347	0
P2	4	-1.112	5.6	-1.106	0.154	0.54	-1.112	0.516	0
P3	6	-1.487	21.7	-1.471	0.303	1.08	-1.468	0.682	1.28
P4	8	-1.992	79.5	-1.971	0.591	1.05	-1.971	0.792	1.05
P5	10	-2.786	162.1	-2.661	0.724	4.49	-2.6914	0.998	3.40
P6	12	-3.785	729.3	-3.662	1.371	3.25	-3.699	1.947	2.27
P7	14	-5.211	1680.9	-4.997	4.274	4.11	-5.106	3.042	2.01
P8	16	-8.633	3600	-8.012	5.291	7.19	-8.475	5.708	1.83
P9	18	-	-	-9.547	7.264	-	-9.827	8.066	-
P10	20	-	-	-11.287	11.919	-	-11.531	13.281	-
Average		-	785.0875	-	3.1921	2.171		3.5379	1.184

| Show Table

DownLoad: CSV

Table 3. Computational results of GA and IWO in the large-scale problems

0	0.1	0.2	0.3	0.4	0.5	0.6	0.7	0.8	0.9	1	Risk aversion factor
8.24	7.39	5.99	4.89	4.98	1.53	1.24	0.83	0.82	0.33	0.21	Return	GA
38.09	15.3	7.9	5.39	5.18	2.16	2.07	1.59	1.6	1.59	1.58	Risk	GA
10.02	8.52	7.36	6.62	6.45	3.1	2.94	2.37	2.35	1.87	1.78	Return	IWO
37.1	14.58	7.75	5.5	5.01	2.03	1.95	1.6	1.53	1.65	1.54	Risk	IWO

| Show Table

DownLoad: CSV

Table 4. Summarized results of IWO algorithm for the deterministic portfolio selection problem with 10 products

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.29630	0.69760	-56.99680	0.95800
0.1	1.19926	0.60310	-50.19030	1.39730
0.2	1.17887	0.49650	-40.04700	1.05580
0.3	1.08299	0.46870	-32.99290	0.95450
0.4	1.06606	0.44740	-26.26950	1.00600
0.5	1.02268	0.44230	-18.88370	0.79310
0.6	0.96123	0.43990	-11.05970	0.96800
0.7	0.87897	0.43960	-4.22660	0.85980
0.8	0.87608	0.43850	2.91500	1.02450
0.9	0.80853	0.42070	8.71360	0.92530
1	0.80043	0.39650	14.14370	1.03380
average	1.01558	0.48098	-19.53584	0.99783

| Show Table

DownLoad: CSV

Table 5. Summarized results of IWO algorithm for the robust portfolio selection problem with 10 products ($\rho = 0.2$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.21507	0.66178	-51.84080	0.97237
0.1	1.20126	0.62716	-45.59330	1.44621
0.2	1.09569	0.51388	-35.16300	1.07660
0.3	1.09358	0.49401	-26.80290	1.07190
0.4	1.05358	0.49130	-19.26790	1.10600
0.5	1.01167	0.49102	-15.62370	0.88859
0.6	0.94073	0.48330	-8.05990	1.06238
0.7	0.90638	0.47573	0.99940	0.94625
0.8	0.86803	0.46300	9.80500	1.15400
0.9	0.80883	0.45101	15.94460	1.22232
1	0.79322	0.40662	19.17160	1.06134
Average	0.99891	0.50535	-14.22099	1.09163

| Show Table

DownLoad: CSV

Table 6. Summarized results of IWO algorithm for the robust portfolio selection problem with 10 products ($\rho = 0.4$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.26527	0.69400	-50.32180	1.05357
0.1	1.20462	0.64976	-45.06690	1.68514
0.2	1.13801	0.59878	-36.04860	1.13963
0.3	1.07096	0.55574	-24.82890	1.01938
0.4	0.99564	0.53676	-18.00080	1.27179
0.5	0.93429	0.53060	-9.88430	1.01640
0.6	0.84920	0.51546	-5.03810	1.06400
0.7	0.83977	0.48991	-0.62960	1.10240
0.8	0.75594	0.44834	10.17900	1.23596
0.9	0.72847	0.39742	16.83960	1.04254
1	0.72423	0.39170	20.17070	1.20169
Average	0.95513	0.52804	-12.96634	1.16659

| Show Table

DownLoad: CSV

Table 7. Summarized results of IWO algorithm for the robust portfolio selection problem with 10 products ($\rho = 0.6$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.22255	0.76641	-45.06180	1.24743
0.1	1.18229	0.72025	-37.85050	1.72660
0.2	1.09349	0.72213	-32.92260	1.47810
0.3	0.99892	0.64398	-19.87430	1.36637
0.4	0.96053	0.64205	-10.73980	1.26848
0.5	0.92569	0.57386	-1.75830	1.01640
0.6	0.90423	0.57049	3.98354	1.06400
0.7	0.85309	0.56688	1.54040	1.10240
0.8	0.78659	0.45533	16.30550	1.21371
0.9	0.75671	0.43173	25.85560	1.27669
1	0.75242	0.41958	28.11682	1.45525
Average	0.94877	0.59206	-6.58231	1.29231

| Show Table

DownLoad: CSV

Table 8. Summarized results of IWO algorithm for the deterministic portfolio selection problem with 50 products

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.4327	0.6733	-32.3022	0.9262
0.1	1.3333	0.6593	-29.6012	1.0732
0.2	1.2003	0.5725	-23.4824	1.0453
0.3	1.1757	0.5570	-19.8822	1.1412
0.4	1.1636	0.5561	-15.9330	1.1919
0.5	1.1184	0.5254	-11.2173	0.7735
0.6	1.0364	0.4975	-6.4951	0.9450
0.7	1.0268	0.4714	-2.2812	0.7081
0.8	0.9291	0.4580	1.6748	1.0726
0.9	0.9080	0.4226	4.6130	1.0580
1	0.8963	0.4052	7.7444	0.9811
Average	1.1110	0.5271	-11.5602	0.9924

| Show Table

DownLoad: CSV

Table 9. Summarized results of IWO algorithm for the robust portfolio selection problem with 50 products ($\rho = 0.2$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.4913	0.7941	-29.2069	1.1059
0.1	1.4243	0.7584	-25.6067	1.1246
0.2	1.3192	0.6295	-19.4184	1.1503
0.3	1.1959	0.5930	-14.7174	1.2152
0.4	1.1603	0.5929	-12.8385	1.3426
0.5	1.1007	0.5813	-4.1561	1.0164
0.6	1.0427	0.5296	1.6449	1.0640
0.7	0.9459	0.4873	3.8428	1.1024
0.8	0.9456	0.4772	6.3238	1.1423
0.9	0.9126	0.4655	10.6742	1.2695
1	0.9099	0.4435	10.7638	1.1976
Average	1.1317	0.5775	-6.6086	1.1573

| Show Table

DownLoad: CSV

Table 10. Summarized results of IWO algorithm for the robust portfolio selection problem with 50 products ($\rho = 0.4$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.4954	0.8165	-27.1532	1.0138
0.1	1.3815	0.7981	-25.3369	1.1246
0.2	1.2973	0.7045	-18.3760	1.1505
0.3	1.1837	0.6121	-13.7562	1.2547
0.4	1.1388	0.6095	-9.9320	1.2114
0.5	1.0932	0.5819	-3.1832	1.0164
0.6	0.9935	0.5793	-0.4791	1.0640
0.7	0.9866	0.5776	2.8418	1.1024
0.8	0.9594	0.5133	10.7808	1.3515
0.9	0.9504	0.5106	12.6253	1.2808
1	0.8817	0.4496	11.7754	1.0439
Average	1.1238	0.6139	-5.4721	1.1467

| Show Table

DownLoad: CSV

Table 11. Summarized results of IWO algorithm for the robust portfolio selection problem with 50 products ($\rho = 0.6$)

Risk aversion factor	Total return	Total risk	Objective function Value	Execution time
0	1.4189	0.8471	-26.1382	1.0433
0.1	1.3362	0.7674	-23.4752	1.1246
0.2	1.2420	0.7352	-19.2514	1.3819
0.3	1.1947	0.7170	-12.8606	1.3767
0.4	1.1767	0.7040	-8.8690	1.3882
0.5	1.1766	0.6931	-2.0909	1.0164
0.6	1.1491	0.6668	1.5295	1.0640
0.7	1.0411	0.6338	2.3588	1.1024
0.8	1.0095	0.5640	6.8008	1.1792
0.9	0.9435	0.5015	11.6470	1.2315
1	0.9186	0.4455	16.9694	1.6090
Average	1.1461	0.6614	-4.8527	1.2288

| Show Table

DownLoad: CSV

Related Papers

Cited by

References

[1]	A. Ben-Tal and A. Nemirovski, Selected topics in robust convex optimization, Mathematical Programming, 112 (2008), 125-158. doi: 10.1007/s10107-006-0092-2.
[2]	A. Ben-Tal, L. El Ghaoui and A. Nemirovski, Robust Optimization, Princeton University Press, 2009. doi: 10.1515/9781400831050.
[3]	D. Bertsimas and M. Sim, The price of robustness, Operations Research, 52 (2004), 35-53. doi: 10.1287/opre.1030.0065.
[4]	D. Bertsimas and D. Pachamanova, Robust multiperiod portfolio management in the presence of transaction costs, Computers & Operations Research, 35 (2008), 3–17. doi: 10.1016/j.cor.2006.02.011.
[5]	R. N. Cardozo and D. K. Smith Jr, Applying financial portfolio theory to product portfolio decisions: An empirical study, The Journal of Marketing, (1983), 110–119.
[6]	L. Cruz, E. Fernandez, C. Gomez, G. Rivera and F. Perez, Many-objective portfolio optimization of interdependent projects with'a priori'incorporation of decision-maker preferences, Applied Mathematics & Information Sciences, 8 (2014), 1517–1526.
[7]	T. Cura, Particle swarm optimization approach to portfolio optimization, Nonlinear Analysis: Real World Applications, 10 (2009), 2396-2406. doi: 10.1016/j.nonrwa.2008.04.023.
[8]	L. Czaplewski, R. Bax, M. Clokie, M. Dawson, H. Fairhead, V.A. Fischetti, S. Foster, B.F. Gilmore, R.E. Hancock and D. Harper, Alternatives to antibioticsa pipeline portfolio review, The Lancet Infectious Diseases, 16 (2016), 239-251.
[9]	S. El-Bizri and N. Mansour, Metaheuristics for Portfolio Optimization, in: International Conference in Swarm Intelligence, Springer, (2017), 77–84.
[10]	H. N. Esfahani, M. h. Sobhiyah and V. R. Yousefi, Project portfolio selection via harmony search algorithm and modern portfolio theory, Procedia - Social and Behavioral Sciences, 226 (2016), 51-58.
[11]	A. Fernández and S. Gómez, Portfolio selection using neural networks, Computers & Operations Research, 34 (2007), 1177–1191.
[12]	R. Fernandes, J. B. Gouveia and C. Pinho, Product mix strategy and manufacturing flexibility, Journal of Manufacturing Systems, 31 (2012), 301-311.
[13]	A. Goli and S. M. R. Davoodi, Coordination policy for production and delivery scheduling in the closed loop supply chain, Production Engineering, 1 (2018), 1-11.
[14]	A. Goli, A. Aazami and A. Jabbarzadeh, Accelerated cuckoo optimization algorithm for capacitated vehicle routing problem in competitive conditions, International Journal of Artificial Intelligence, 16 (2018), 88-112.
[15]	N. Gülpnıar and E. Çanakoğlu, Robust portfolio selection problem under temperature uncertainty, European Journal of Operational Research, 256 (2017), 500-523. doi: 10.1016/j.ejor.2016.05.046.
[16]	M. A. Lejeune and S. Shen, Multi-objective probabilistically constrained programs with variable risk: Models for multi-portfolio financial optimization, European Journal of Operational Research, 252 (2016), 522-539. doi: 10.1016/j.ejor.2016.01.039.
[17]	C.-C. Lin and Y.-T. Liu, Genetic algorithms for portfolio selection problems with minimum transaction lots, European Journal of Operational Research, 185 (2008), 393-404.
[18]	L. L. Macedo, P. Godinho and M. J. Alves, Mean-semivariance portfolio optimization with multiobjective evolutionary algorithms and technical analysis rules, Expert Systems with Applications, 79 (2017), 33-43.
[19]	H. Markowitz, Portfolio selection, The Journal of Finance, 7 (1952), 77-91.
[20]	H. Markowitz, Portfolio Selection: Efficient Diversification of Investments, Cowles Foundation Monograph, No. 16, John Wiley & Sons, 1959.
[21]	H. M. Markowitz and G. P. Todd, Mean-variance Analysis in Portfolio Choice and Capital Markets, John Wiley & Sons, 2000.
[22]	D. Maringer and P. Parpas, Global optimization of higher order moments in portfolio selection, Journal of Global Optimization, 43 (2009), 219-230. doi: 10.1007/s10898-007-9224-3.
[23]	A. R. Mehrabian and C. Lucas, A novel numerical optimization algorithm inspired from weed colonization, Ecological Informatics, 1 (2006), 355-366. doi: 10.1016/B978-0-12-416743-8.00001-4.
[24]	M. Montajabiha, A. Arshadi Khamseh and B. Afshar-Nadjafi, A robust algorithm for project portfolio selection problem using real options valuation, International Journal of Managing Projects in Business, 10 (2017), 386-403.
[25]	J. M. Mulvey, R. J. Vanderbei and S. A. Zenios, Robust optimization of large-scale systems, Operations Research, 43 (1995), 264-281. doi: 10.1287/opre.43.2.264.
[26]	G. V. Pai and T. Michel, Metaheuristic optimization of constrained large portfolios using hybrid particle swarm optimization, International Journal of Applied Metaheuristic Computing, 8 (2017), 1-23.
[27]	A. Ponsich, A. L. Jaimes and C. A. C. Coello, A survey on multiobjective evolutionary algorithms for the solution of the portfolio optimization problem and other finance and economics applications, IEEE Transactions on Evolutionary Computation, 17 (2013), 321-344.
[28]	B. Qu, Q. Zhou, J. Xiao, J. Liang and P. Suganthan, Large-scale portfolio optimization using multiobjective evmolutionary algorithms and preselection methods, Mathematical Problems in Engineering, 17 (2017), 1-14.
[29]	S. M. Seyedhosseini, M. J. Esfahani and M. Ghaffari, A novel hybrid algorithm based on a harmony search and artificial bee colony for solving a portfolio optimization problem using a mean-semi variance approach, Journal of Central South University, 23 (2016), 181-188.
[30]	F. Solatikia, E. Kiliç and G. W. Weber, Fuzzy optimization for portfolio selection based on embedding theorem in fuzzy normed linear spaces, Organizacija, 47 (2014), 90-97.
[31]	A. L. Soyster, Convex programming with set-inclusive constraints and applications to inexact linear programming, Operations Research, 21 (1973), 1154-1157. doi: 10.1287/opre.22.4.892.
[32]	M. A. Takami, R. Sheikh and S. S. Sana, Product portfolio optimisation using teaching learning-based optimisation algorithm: a new approach in supply chain management, International Journal of Systems Science: Operations & Logistics, 3 (2015), 236–246.
[33]	E. B. Tirkolaee, A. Goli, M. Bakhsi and I. Mahdavi, A robust multi-trip vehicle routing problem of perishable products with intermediate depots and time windows, Numerical Algebra, Control & Optimization, 7 (2017), 417–433. doi: 10.3934/naco.2017026.
[34]	M. Tuba and N. Bacanin, Upgraded firefly algorithm for portfolio optimization problem, in: Computer Modelling and Simulation, 2014 UKSim-AMSS 16th International Conference on, IEEE, (2014), 113–118.