Hybridizing a fuzzy multi-response Taguchi optimization algorithm with artificial neural networks to solve standard ready-mixed concrete optimization problems

Creative Commons License

ŞİMŞEK B., İÇ Y. T., ŞİMŞEK E. H.

INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS, cilt.9, sa.3, ss.525-543, 2016 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 9 Sayı: 3
  • Basım Tarihi: 2016
  • Doi Numarası: 10.1080/18756891.2016.1175816
  • Dergi Adı: INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.525-543
  • Anahtar Kelimeler: Standard ready-mixed concrete, Multi-response optimization, Taguchi method, Fuzzy TOPSIS, Artificial neural networks, RECYCLED AGGREGATE CONCRETE, SELF COMPACTING CONCRETE, COMPRESSIVE STRENGTH, DECISION-MAKING, OPTIMAL MIXTURE, SILICA FUME, FLY-ASH, DESIGN, TOPSIS, PROPORTIONS
  • Ankara Üniversitesi Adresli: Evet

Özet

In this study, a fuzzy multi-response standard ready-mixed concrete (SRMC) optimization problem is addressed. This problem includes two conflicting quality optimization objectives. One of these objectives is to minimize the production cost. The other objective is to assign the optimal parameter set of SRMC's ingredient to each activity. To solve this problem, a hybrid fuzzy multi-response optimization and artificial neural network (ANN) algorithm is developed. The ANN algorithm is integrated into the multi-response SRMC optimization framework to predict and improve the quality of SRMC. The results show that fuzzy multi-response optimization model is more effective than crisp multi-response optimization model according to final production cost. However, the ANN model also gave more accurate results than the fuzzy model considering the regression analysis results.