ABSTRACT

Stochastic bid price optimization of truckload carriers in simultaneous independent transportation auctions requires estimating the probability distributions of the clearing prices. Historical data can be used for this purpose. The sole estimation method found in the literature for this problem setting only takes the length of the auction load into account. In this paper, we devise methods for load-specific parameter estimation by filtering out data coming from past auction loads with distant origin or destination locations. Through simulations, we demonstrate that our methods can improve the profitability of the carriers compared to the previously used method.

摘要

在同时进行的独立运输拍卖中，对货车承运人进行随机竞价优化需要估计清算价格的概率分布。历史数据可用于此目的。文献中针对该问题设置的唯一估计方法仅考虑了拍卖负载的长度。在本文中，我们设计了一种方法，通过滤除来自具有远距离始发地或目的地位置的过去拍卖负荷的数据来进行负荷特定参数估计。通过仿真，我们证明了与以前使用的方法相比，我们的方法可以提高运营商的盈利能力。