An ambulance location problem for covering inherently rare and random road crashes. (January 2021)

• Record Type: Journal Article
• Authors: Mohri, Seyed Sina; Haghshenas, Hossein
• Is Part Of: Computers & industrial engineering. Volume 151(2021)
• Publication Date: 2021-01
• Subjects: Edge covering problem -- Empirical Bayes method -- Ambulance location problem -- Road crashes; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2020.106937
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: The ambulance location problem is modeled with an edge covering location model. The network demand includes both frequency and severity of rare and random crashes. The rareness in the frequency of crashes is inserted by placing crashes on edges. Empirical Bayes is used to address randomness in the frequency of crashes. Interesting insights are suggested by testing the model on a real case study. Abstract: Ambulance Location Problem (ALP) is studied in the field of covering road crashes, which are often rare and random in nature. On the account of rareness, crashes should be attributed to the network edges rather than nodes. Therefore, a covering problem with edge demand is the most compatible location problem. Accordingly, this paper proposes an edge maximal covering location problem with partial coverage of the facilities on the edges. Randomness is a feature associated with the frequency and severity of crashes. An Empirical Bayes (EB) method is applied to the observed frequency of crashes to reduce errors caused by the random feature of the crashes. The randomness associated with crash severity is investigated by adding the data derived from Property Damage Only (PDO) crashes to the network demand. As a result, an equivalent PDO measure is proposed to satisfy the demand of the network edges. The proposed method is applied to a real case study and some insights are presented.