Fraud in public procurement processes is a scourge in all states and societies that produces great economic losses ultimately borne by citizens. The first step in the fight against fraud is measurement; however, it is complex to estimate existing fraud because in order to be measured it must first be identified. It is estimated that detected fraud is only the tip of the iceberg of all existing fraud, as the following graph on the different levels of fraud shows.
The following three key pillars for an effective fight against procurement fraud are proposed:
- Source Integration: in order to facilitate the interoperability of systems and accessibility of data for obtaining legitimate information.
- Investigative System: enables investigators to work with a tool through which they can set up automated alerts, conduct fraud investigation and manage cases.
- Action Plan: allows the establishment of the methodology, objectives and actions in the prevention and detection of fraud, together with the monitoring and review of objectives.
Source Integration
Thanks to efforts in recent years to facilitate transparency in procurement processes, states are now providing public procurement portals where the various central, regional and local administrations must carry out the processing. This is the case of the Public Sector Procurement Platform for the Spanish state, which covers state, autonomous community and local government procurement either directly or by connecting to the individual platforms.
From these portals we can obtain data on the estimated value of a contract, the contracting entity, the type of process, relevant dates, information on tenders and the bodies involved in the process.
In addition to the information on the contracting process, it is important to have data on the bidding companies, usually available in government databases such as the Official Register of Bidders in Spain or through private companies such as eInforma. From these data sources we can extract information about the value of the company, financial solvency, number of employees, management and administrative bodies.
For a complete analysis, information can be extracted about the people who have been directly involved in the recruitment process through their profiles in the main social networks, with the aim of establishing links between different people as well as between companies and individuals, such as the popular LinkedIn network.
Red Flags
Based on a history with the information gathered from the sources described above, it will be necessary to identify Red Flags. Red flags are signs that can alert of the existence of a possible fraud. These red flags can consist of the simple identification of thresholds as extreme values or in the advanced identification of relationships between the different actors, such as conflicts of interest that occur in the connection of public personnel and employees of the bidding companies or collusion processes between several of these companies.
The flags can be configured so that they warn of risk situations at different points in the contracting process, basically at the time when the process is advertised, immediately after the opening of the envelopes of the bids and finally at the award of the contract.
Research system
Once the information is accessible, it will be necessary to have a system that has at least the following components
- Asystem of rules that enables the coding of red flags and their weighting to alert to situations that can be investigated.
- Intelligence graph, which allows the investigation of possible fraudulent cases through consultations and navigation, created through the existing relationship between the different actors involved. This will help the investigator in generating hypotheses, capturing evidence and issuing a report on the outcome of the investigation that may lead to legal action.
- Unified management of cases, allowing the assignment of cases, storage of evidence, monitoring and evolution of the cases investigated in their different states, as a collaborative platform for the different investigators.
- Control panel for monitoring and following up objectives with metrics and aggregate information on investigation cases, allowing an overview of the degree of compliance with these objectives.
Plan of action
The action plan will define the general objectives and lines of action to improve the prevention and detection of fraud by monitoring companies or individuals involved in the award of contracts susceptible to fraud. The general objectives should be broken down into specific, measurable and time-bound objectives, which should be monitored and followed up throughout the action period.
Finally
From Future Space S.A. we want to contribute with our knowledge and experience to facilitate the fraud detection in the public procurement processes through the Smartgraph application that has the mentioned components in a unified way and additionally it is provided with the capacity to complement the fraud detection through automatic learning algorithms. Additionally for the collection and analysis of open sources FS OSINT offers an easy way to collect data from social networks, press, blogs and any source of public information on the Internet.
