Open AccessNeural Network Model for Design Constructability Assessment
Author(s) -
Rosli Mohamad Zin,
Muhd Zaimi Abd Majid,
Che Wan Fadhil Che Wan Putra,
Abdul Hakim Mohammed
Publication year - 2012
Publication title -
jurnal teknologi/jurnal teknologi
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.191
H-Index - 22
eISSN - 2180-3722
pISSN - 0127-9696
DOI - 10.11113/jt.v40.404
Subject(s) - artificial neural network , computer science , humanities , artificial intelligence , philosophy
Kertas ini memaparkan satu kajian mengenai penilian kebolehbinaan reka bentuk menggunakan kaedah rangkaian neural tiruan (ANN). Model neural timbalbalik berbilang lapis yang dihasilkan mengandungi 12 pemboleh ubah input dan 1 pemboleh ubah output. Pemboleh ubah input terdiri dari tahap gunapakai faktor-faktor kebolehbinaan iaitu merupakan subfaktor prinsip-prinsip terpenting kebolehbinaan fasa reka bentuk manakala pemboleh ubah output adalah tahap kebolehbinaan reka bentuk. Pembangunan model dibuat melalui lima peringkat: mengenal pasti prinsip-prinsip kebolehbinaan fasa reka bentuk, menentukan darjah kepentingan prinsip-prinsip kebolehbinaan, menghasilkan satu kerangka untuk mengukur tahap guna pakai prinsip-prinsip kebolehbinaan dan tahap reka bentuk, mengumpul data-data projek terdahulu, dan mengguna pakai kaedah ANN untuk menilai kebolehbinaan reka bentuk. Setiap peringkat pembentukan model adalah diterangkan. Data-data projek terdahulu berkaitan pembinaan rasuk telah dikumpulkan dari kontraktor-kontraktor yang mempunyai pengalaman beberapa tahun dalam pembinaan bangunan. Sejumlah 78 set data telah digunakan untuk melatih dan menguji rangkaian neural. Penentuan bilangan optima bagi nod tersembunyi, aras tersembunyi, pemberat awalan bagi penghubung antara nod, dan bilangan iterasi latihan adalah berdasarkan cubaan dan ralat. Artikek rangkaian neural terbaik didapati mengandungi 12 nod input, 5 nod tersembunyi dan 1 nod output. Kata kunci: Rangkaian neural tiruan (ANN), kebolehbinaan reka bentuk, model neural timbalbalik berbilang lapis, faktor, faktor-faktor kebolehbinaan This paper presents an artificial neural network (ANN) technique of analysis for the assessment of design constructability. The multilayer back-propagation neural network model consists of 12 and 1 output variable. The input variables are the level of applications of constructability factors, which are sub-factors of the most important design phase constructability principles while the output variable is the level of design constructability. The development of the model goes through five main stages: identifying the design phase constructability principles, identifying the degree of importance of the constructability principles, formulating a framework for measuring the level of application of constructability principles and design constructability, collecting historical project data, and applying ANN to assess design constructability. Each stage of the model development is described. Historical project data sets related to beam construction have been collected from various constructors that have at least several years of experience in building construction. A total of 78 data sets were used to test and train the network. The determination of the optimum number of hidden nodes, hidden layers, initial weights of the links connecting the nodes, and the number of epochs for training the networks, are normally based on trial and error. The best architecture was found to consist of 12 input nodes, 5 hidden nodes, and 1 output node. Key words: Artificial neural network (ANN), design constructability, multilayer back-propagation neural network model, constructability factors