FIS-SMED: a fuzzy inference system application for plastic injection mold changeover

Karasu M. K., SALUM L.

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY, cilt.94, ss.545-559, 2018 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 94
  • Basım Tarihi: 2018
  • Doi Numarası: 10.1007/s00170-017-0799-7
  • Dergi Adı: INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.545-559
  • Anahtar Kelimeler: Fuzzy inference system, SMED, Changeover, Parameter setting, ARTIFICIAL NEURAL-NETWORK, RESPONSE-SURFACE METHODOLOGY, WARPAGE OPTIMIZATION METHOD, THIN-WALL PARTS, PROCESS PARAMETERS, TAGUCHI METHOD, GENETIC ALGORITHM, REDUCING WARPAGE, PROCESS DESIGN, MECHANICAL-PROPERTIES
  • Dokuz Eylül Üniversitesi Adresli: Evet

Özet

Dr. Shingo's SMED (single minute exchange of dies) methodology is the most well-known method for changeover time reduction using both simple methodological solutions and tool/design changes. Simplification and standardization are the main technique of SMED to make the changeover process independent from personal experience. However, in plastic injection molding, process parameter setting after changing the molds totally depends on the varied expertise levels of setup experts. The number of available setup experts in a shift dictates the number of changeover that can be given to the production plan. Due to this dependency, expected benefits of SMED cannot be realized. In this paper, an application of a fuzzy inference system (FIS) is presented for parameter adjustments during changeovers on plastic injection molds. The proposed system captures the highest level of domain expertise and makes it applicable by machine operators. Integrating this system into SMED applications encourages production lot size reduction. Moreover, proposed FIS increases the quality awareness of machine operators and can be used to train new ones.