A double auction based mathematical market model and heuristics for internet-based secondhand durable good markets


Özer A. H.

COMPUTERS & OPERATIONS RESEARCH, cilt.111, ss.116-129, 2019 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 111
  • Basım Tarihi: 2019
  • Doi Numarası: 10.1016/j.cor.2019.06.005
  • Dergi Adı: COMPUTERS & OPERATIONS RESEARCH
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.116-129
  • Anahtar Kelimeler: Electronic market, Double auction, Budget constraint, Used good trading, Secondhand market, Market design, Mathematical modeling, Integer programming, Combinatorial optimization, EMPIRICAL-ANALYSIS, CIRCULAR ECONOMY, EFFICIENCY, EXCHANGES, TRADE
  • Marmara Üniversitesi Adresli: Evet

Özet

Trading of used goods in secondhand markets is considered to have both environmental and economic benefits. For the trading of used goods as well as new ones, this study proposes a periodic double auction model for Internet-based electronic markets whereby each bidder can place both asks and bids for selling and purchasing goods respectively during a trading period. The proposed model allows each bidder to place a spending limit order such that the difference between the cost of the purchased goods and the income obtained from the sold goods does not exceed this limit. Furthermore, since a bidder may be indifferent to multiple goods, e.g. copies of a same title, the model also provides a mechanism so that bidders may combine a number of bids inside a set and put a limit on the number of goods to be purchased in this set. The model is mathematically defined, and the corresponding winner determination problem is formulated using linear integer programming. Since the problem is shown to be NP-hard, a number of heuristic methods are also proposed. Performances of these methods are evaluated on a comprehensive test suite and statistical analyses of the results are presented. Furthermore, possible economical contribution of the model is also evaluated. The results indicate that the proposed model can be used efficiently in large-scale markets with tens of thousands of bidders. (C) 2019 Elsevier Ltd. All rights reserved.