Prior-free auction mechanism for online supplier with risk taking. (July 2019)

• Record Type: Journal Article
• Authors: Ding, Lili; Liu, Mengxi; Kang, Wanglin; Zhao, Xin
• Is Part Of: Computers & industrial engineering. Volume 133(2019)
• Page Start: 1
• Page End: 8
• Publication Date: 2019-07
• Subjects: Prior-free auction -- Risk management -- Competitive analysis -- Online algorithm; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2019.04.034
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: A novel prior-free auction mechanism for suppliers with incomplete knowledge. A competitive risk analysis framework by introducing risk and forecast is presented. Enable a supplier to utilize and exploit forecast under different risk tolerances. Abstract: Auction is a useful trade manner for digital goods procurements in e-market places. In this paper, we present a novel prior-free auction mechanism for supplier that serves online buyers. As the recommended the auction mechanism, the supplier has incomplete knowledge of the distribution of buyers' valuations and makes a decision to accept or to reject each bid before each buyer leaves. To analyze auction performance without making assumptions about the prior input distribution, the competitive analysis of online algorithmic mechanism is motivated. We extend the original model to present a competitive risk analysis framework by introducing risk and forecast. Moreover, a λ -tolerance algorithmic mechanism and an n -phase λ -tolerance algorithmic mechanism are proposed. They enable a supplier to utilize and exploit forecast under different risk tolerances. Solutions of these risk algorithms yield many insights as follows: a supplier's risk management performance, i.e., the restricted competitive ratio can be improved by putting reasonable forecasts comparing with the traditional competitive analysis; a supplier can also control his risk behavior even if the forecast is incorrect; this risk framework provides a smooth generalization for a supplier's riskless choices and risky choices. .