The aim of automating the task of generating flow control designs for oil and gas flow comes with the different dimension constraints and cost factors. The resulted designs should satisfy given dimension boundaries and pre- specified conditions. To make the module more efficient and automate we combine the machine learning and constraint satisfaction module which resulted in reduction of time complexity and how the accuracy gets maintained. The result shows how the separate module of machine learning and optimization module work and how the results get vary when we combine both modules. The constraint we want to optimize are size and cost of the design. The main factors we considered for measuring performance are time complexity and accuracy.

Constraint Satisfaction Problem, Constraint Optimization, Optimization Engine, Machine Learning

[1] Vincent Tam and K.T. Ma, “Optimizing Personal Computer Configurations with Heuristic-Based Search Methods,” 129–140, 2012.
[2] Igor Rivin and Ramin Zabih “An Algebraic Approach to Constraint Satisfaction Problems,” 2011.
[3] Guido Tack, Tias Gunsand, “Introduction to the special issue on Combining Constraint Solving with Mining and Learning,” March 2017 Artificial Intelligence 244:1-5.
[4] P.C. Fourie,A.A. Groenwold, “The particle swarm optimization algorithm in size and shape optimization,”2001.
[5] A. Javier Larrosa and B. Rina Dechter, “Boosting Search with Variable Elimination in Constraint Optimization and Constraint Satisfaction Problems,” MIT Press, 2007.
[6] Luc De Raedt1, Siegfried Nijssen2, Barry O’Sullivan3 , and Pascal Van Hentenryck4, “Constraint Programming meets Machine Learning and Data Mining,” Report from Dagstuhl Seminar 11201.
[7] Hana Rudová “Constraint Satisfaction with Preferences,” IEEE, 2014.
[8] L. De Raedt, T. Guns, and S. Nijssen ,“Constraint Programming for Itemset Mining In ACM SIGKDD”, Int. Conf. KDD’08, Las Vegas, Nevada, USA, 2008.
[9] L. De Raedt, T. Guns, and S. Nijssen, “Constraint programming for data mining and machine learning,” In Twenty-Fourth AAAI Conference on Artificial Intelligence (AAAI10), pages 1671–1675, 2010.
[10] M. Khiari, P. Boizumault, and B. Crémilleux, “Local constraint-based mining and set constraint programming for pattern discovery”, ECML/PKDD-09 Workshop, pages 61–76, Bled, Slovenia, 2009.
[11] M. Khiari, P. Boizumault, and B. Crémilleux “Constraint programming for mining n-ary patterns,” In 16th Int. Conf. on Principles and Practice of Constraint Programming (CP’10), volume 6308 of LNCS, pages 552–567. Springer, 2010.
[12] A. Javier Larrosa, B. Rina Dechter, “Boosting Search with Variable Elimination in Constraint Optimization and Constraint Satisfaction Problems,” 2002 Kluwer Academic Publishers.
[13] Frost, D. and R. Dechter: 1994, “In Search of the Best Constraint Satisfaction Search,” In: Proceedings of the 12th AAAI. pp. 301–306.
[14] Larrosa, J.: 2000, “Boosting Search with Variable Elimination,” In: Proc. of the 6th CP,Singapore, pp. 291–305.
[15] Dechter, R., K. Kask, and J. Larrosa: 2001, “A General Scheme for Multiple Lower Bound Computation in Constraint Optimization,” In: Proc. of the 7th CP. pp. 346–360.
[16] Meseguer, P., J. Larrosa, and M. Sanchez: 2001, “Lower Bounds for Non-Binary Constraint Optimization Problems,” In: Proc. of the 7th CP.pp.3