Ron Triepels

Uncovering Document Fraud in Maritime Freight Transport Based on Probabilistic Classification

Deficient visibility in global supply chains causes significant risks for the customs brokerage practices of freight forwarders. One of the risks that freight forwarders face is that shipping documentation might contain document fraud and is used to declare a shipment. Traditional risk controls are ineffective in this regard since the creation of shipping documentation is uncontrollable by freight forwarders. In this paper, we propose a data mining approach that freight forwarders can use to detect document fraud from supply chain data. More specifically, we learn models that predict the presence of goods on an import declaration based on other declared goods and the trajectory of the shipment. Decision rules are used to produce miscoding alerts and smuggling alerts. Experimental tests show that our approach outperforms the traditional audit strategy in which random declarations are selected for further investigation.

Data-driven fraud detection in international shipping.

Document fraud constitutes a growing problem in international shipping. Shipping documentation may be deliberately manipulated to avoid shipping restrictions or customs duties. Well-known examples of such fraud are miscoding and smuggling. These are cases in which the documentation of a shipment does not correctly or entirely describe the goods in transit. In an attempt to reduce the risks of document fraud, shipping companies and customs authorities typically perform random audits to check the accompanying documentation of shipments. Although these audits detect many fraud schemes, they are quite labor intensive and do not scale to the massive amounts of cargo that is shipped each day. This paper investigates whether intelligent fraud detection systems can improve the detection of miscoding and smuggling by analyzing large sets of historical shipment data. We develop a Bayesian network that predicts the presence of goods on the cargo list of shipments. The predictions of the Bayesian network are compared with the accompanying documentation of a shipment to determine whether document fraud is perpetrated. We also show how a set of discriminative models can be derived from the topology of the Bayesian network and perform the same fraud detection task. Our experimental results show that intelligent fraud detection systems can considerably improve the detection of miscoding and smuggling compared to random audits.

Anomaly Detection in Real-Time Gross Settlement Systems.

We discuss how an autoencoder can detect system-level anomalies in a real-time gross settlement system by reconstructing a set of liquidity vectors. A liquidity vector is an aggregated representation of the underlying payment network of a settlement system for a particular time interval. Furthermore, we evaluate the performance of two autoencoders on real-world payment data extracted from the TARGET2 settlement system. We do this by generating different types of artificial bank runs in the data and determining how the autoencoders respond. Our experimental results show that the autoencoders are able to detect unexpected changes in the liquidity flows between banks.

Detection and explanation of anomalous payment behavior in real-time gross settlement systems

In this paper, we discuss how to apply an autoencoder to detect anomalies in payment data derived from an Real-Time Gross Settlement system. Moreover, we introduce a drill-down procedure to measure the extent to which the inflow or outflow of a particular bank explains an anomaly. Experimental results on real-world payment data show that our method can detect the liquidity problems of a bank when it was subject to a bank run with reasonable accuracy.

A Comparison of Three Models to Predict Liquidity Flows between Banks Based on Daily Payments Transactions

The analysis of payment data has become an important task for operators and overseers of financial market infrastructures. Payment data provide an accurate description of how banks manage their liquidity over time. In this paper we compare three models to predict future liquidity flows from payment data: 1) a moving average model, 2) a linear dynamic system that links the inflow of banks with their outflow, and 3) a similar dynamic system but with a constraint that guarantees the conservation of liquidity. The error graphs of one-step-ahead predictions on real-world payment data reveal that the moving average model performs best, followed by the dynamic system with constraint, and finally the dynamic system without constraint.

Auditing Data Reliability in International Logistics

Data reliability closely relates to the risk management in international logistics. Unreliable data negatively affect the business in various ways. Due to the competence specialization and cooperation among the business partners in a logistics chain, the business in a focal company is inevitably dependent on external data sources from its partner, which is impractical to control. In this paper, we present a research-in-progress on an analysis method with Bayesian networks. The goal is to support auditorâ??s assessment on the reliability of the external data. A case study is provided to illustrate the merits of Bayesian networks when dealing with the data reliability problem.