A Neural Network Mobile Learning Application For Autonomous Improvement In A Flexible Manufacturing Environment

Siew , Jit Ping (2016) A Neural Network Mobile Learning Application For Autonomous Improvement In A Flexible Manufacturing Environment. PhD thesis, Universiti Sains Malaysia.

[img]
Preview
PDF
Download (411Kb) | Preview

    Abstract

    Kajian ini memberi tumpuan kepada inovasi berasaskan telekomunikasi dan teknologi komputer di kilang pengeluaran "MF" untuk menjanakan pulangan nilai yang lebih tinggi. Process pembuatan moden merupakan industri yang sangat kompetitif, dan kos kerugian daripada kecacatan dalam pengeluaran produk adalah tinggi. Berdasarkan kaji selidik aktiviti process pembuatan, proses percetakan stensil (SPP) telah dipilih sebagai kawasan kajian. Keputusan ini berdasarkan ulasan kesusasteraan yang menunjukkan bahawa sekurang-kurangnya 50% daripada kecacatan dalam pemasangan papan litar bercetak berasal dari SPP, dan data kecacatan sebenar yang dikumpul semasa penyiasatan. Memandangkan persekitaran kerja sambil berdiri oleh krew pengendali mesin yang terus menerus bergerak, cabarannya adalah untuk memberi keupayaan autonomi melalui pengetahuan mengenai prestasi kerja mereka dengan penggunaan aplikasi pembelajaran mudah alih. Untuk mencapai objektif ini, peranti mudah alih dimuatkan dengan sebuah aplikasi Android yang digunakan untuk menyampai maklumat yang diproses oleh algoritma rangkaian neural. This study is focused on how an innovation based on telecommunication and computer technologies at a manufacturing facility “MF” is implemented to generate higher value returns. Modern manufacturing has evolved into a very competitive industry and wastages resulting from process defects are very costly. Based on a survey of the manufacturing floor activities, the stencil printing process (SPP) was selected as the area of research. This decision was based on literature reviews which indicated that at least 50% of defects in the printed circuit board (PCB) assembly originated from SPP, and actual defects data collected during the survey. Given the standing work environment of the machine operators who are continuously on the move, the challenge is therefore, to empower them with knowledge on their performances relative to defects with a mobile learning application, and to stimulate an autonomous process improvement. To attain this objective, a mobile device loaded with an Android app is used to present information that is processed by a neural network algorithm.

    Item Type: Thesis (PhD)
    Subjects: Q Science > QA Mathematics > QA1 Mathematics (General)
    Divisions: Pusat Pengajian Sains Matematik (School of Mathematical Sciences)
    Depositing User: Mr Noorazilan Noordin
    Date Deposited: 02 Mar 2017 15:08
    Last Modified: 02 Mar 2017 15:08
    URI: http://eprints.usm.my/id/eprint/32261

    Actions (login required)

    View Item
    Share