Bitte verwenden Sie diesen Link, um diese Publikation zu zitieren, oder auf sie als Internetquelle zu verweisen: https://hdl.handle.net/10419/317475 
Erscheinungsjahr: 
2021
Quellenangabe: 
[Journal:] Journal of Business Economics and Management (JBEM) [ISSN:] 2029-4433 [Volume:] 22 [Issue:] 2 [Year:] 2021 [Pages:] 320-345
Verlag: 
Vilnius Gediminas Technical University, Vilnius
Zusammenfassung: 
Apart from the loss of time and money, disputes between public authority and private partner in China's public-private partnership (PPP) projects are destroying the government's image of PPP support and the private partner's investment confidence. This article aims to explore the main causes for PPP disputes, present the results of disputes, and then predict the litigation outcomes. Based on 171 PPP litigation cases from China Judgements Online within 2013-2018, the research identified 17 legal factors and explained how these factors influence the litigation outcomes, which are named as "prediction approach" in this study. Nine machine learning (ML) models were trained and validated using the data from 171 cases. The ensemble model of gradient boosting decision tree (GBDT), k-nearest neighbor (KNN) and multi-layer perceptron neural network (MLP) performed best compared with other nine individual ML models, and obtained a prediction accuracy of 96.42%. This study adds meaningful insights to PPP dispute avoidance, such as high compensation of expected revenues could prevent the government from terminating the contract unilaterally. To some extent, if parties consider the case litigation outcome, they are more likely prefer a rational settlement out of court to avoid further aggravation of the dispute, and would also alleviate the pressure of litigation in China.
Schlagwörter: 
public-private partnership (PPP)
project management
dispute causes
court decisions
litigation prediction
artificial intelligence
case study
JEL: 
C53
D74
D80
K41
L32
Persistent Identifier der Erstveröffentlichung: 
Creative-Commons-Lizenz: 
cc-by Logo
Dokumentart: 
Article

Datei(en):
Datei
Größe
516.79 kB





Publikationen in EconStor sind urheberrechtlich geschützt.