Digital tenders and automated procedures: New rules for public procurement
![](https://wp.portolano.it/wp-content/uploads/2024/06/LS-blog-procurement-1-500x263.png)
The Italian version of this article has been published on February 16, 2024 on AgendaDigitale.eu, within our “Legal Health” bi-monthly column.
The digitization of tendering procedures is one of the most important issues addressed by the new Public Procurement Code,[1] which systematically reorganizes and significantly changes the rules on the subject. An entire section of the code is devoted to the subject.[2]
Digitization of the public sector is one of the main objectives of the Italian National Recovery and Resilience Plan (“NRRP”), Mission 1. For tendering procedures, this calls for “establishing modalities for digitizing the procedures for all public contracts and concessions and meeting interoperability and interconnectivity requirements.”[3] It also calls for creation of a national e-procurement system that can interface with the management systems of public administrations.[4]
The objective is the complete digitization of the entire lifecycle of public contracts, understood as all the activities from planning to establishing requirements to full execution. [5] This series of activities is one of the most onerous tasks for administrations in terms of both time and resources. The provision marks a major change, as it requires each phase in the public supply chain to be able to be completed digitally.
2024 changes to digitization of tender procedures
On January 1, 2024, the provisions on the digitization of the procurement lifecycle became fully effective. Changes include operation of the digital procurement ecosystem, or e-procurement, through certified platforms and the National Public Contracts Database (Banca Dati Nazionale dei Contratti Pubblici – “BDNCP”).[6]
The e-procurement rules call for interoperable telematic platforms as the standard means of managing the contract lifecycle, with the aim of creating a single procurement portal.[7]
Provisions on the automation of the public contract lifecycle are also of interest These explicitly mention the possibility of using artificial intelligence (AI) systems to manage tenders.
The new rules will apply to all procedures undertaken on or after January 1, 2024 (including award procedures included in the NRRP); those undertaken on or before December 31, 2023 will follow National Anticorruption Authority (Autorità Nazionale Anticorruzione – “ANAC”) Resolution No. 582 of December 13, 2023.[8]
Certified digital platforms
As of January 1, 2024, contracting stations and economic operators must use certified digital procurement platforms to conduct the entire lifecycles of public contracts, from planning and design to the award and execution of contracts.
By way of example, through these platforms interoperable with the BDNCP, economic operators will be able to access tender documentation, transmit data and documents, and submit the European Single Procurement Document and tenders; contracting stations will be able to request a Tender Identification Number (Codice Identificativo di Gara –“CIG”) and carry out technical, accounting, and administrative checks on contracts during execution.[9]
At present, there are 46 certified digital procurement platforms throughout the country, and some are only operational for certain phases of the contract lifecycle. However, administrations that do not yet have their own platforms can use the platforms made available by other entities, such as other contracting stations, central purchasing bodies and aggregators, and regions and autonomous provinces (which, in turn, can make use of external operators that guarantee that platforms function and are secure).[10] Despite this, many operators are experiencing problems using the new digital platforms for procurement, for example, in the procedure for requesting CIGs.[11]
The national public contracts database
As of January 1, 2024, the new BDNCP—consisting of six sections—is also operational. The BDNCP makes information and services available for effective digital management of the lifecycle of public contracts and ensures compliance with publicity and transparency obligations.
This database has access to (and is fed) the information contained in databases managed by administrations (such as the registry of natural and legal persons, the business registry, the Ministry of Justice, and the Revenue Agency) and databases managed by public companies and public service concessionaires that hold data necessary for the digital lifecycle of contracts. It also interacts with the digital procurement platforms used by individual contracting stations and central purchasing bodies, making the data and information needed to manage procedures available to all interested parties.
The use of automated procedures
A particularly innovative rule is found in the new Article 30 of the code, which covers “use of automated procedures in the lifecycle of public contracts.” Automated decision-making can be understood as all decisions on issues arising during the procurement process that are made by an algorithm according to precise predetermined criteria.
Through this new article, the code seems to support the implementation of such technologies—starting from the tender evaluation phase—by providing their use “where possible” to improve the efficiency and effectiveness of tendering procedures and expressly mentioning technological solutions, such as artificial intelligence. This provision is designed to govern the (near?) future—as the code’s introductory report states—insofar as “at present, in the context of tendering procedures, mostly non-learning algorithms are used for the automatic comparison of certain parameters characterizing the bids that are knowable.”
However, in the near future large quantities of data may become available that make it possible to train learning algorithms so they can be applied to more complex tendering procedures. That makes the inclusion here of material on the principles to be used to govern such systems particularly useful. Article 30 sets forth technical standards that guarantee transparency and security. As one example, contracting stations must make available the source code and any documents necessary to understand the logic underlying an automated system and must introduce tender clauses that ensure that assistance and maintenance necessary to correct errors arising from automation are provided. Automated decisions must be knowable and understandable, and there must always be human input capable of “checking, validating, or disproving” such decisions.
Automated procedures and uses
Article 30 of the new code opens up numerous opportunities for administrations in the area of digitization. It gives the green light to the development of automated systems that also use artificial intelligence, which can potentially be applied to all tendering, pre-award, and post-award activities.
Artificial intelligence systems could assist operators in planning tenders and evaluating offers. There is a wealth of research on the use of these technologies in public procurement.[12] Among the technologies most useful for increasing efficiency in tender management are predictive algorithms that exploit AI systems, especially machine learning (ML) technologies.
What predictive algorithms can do
Predictive algorithms work via ML functions that detect patterns in historical supply chain data, i.e., the critical points and most crucial factors in determining the correct quantity of supplies, such as unpredictable risks, logistics, and optimization.
Using ML systems, a machine learns from data and can conduct analysis without a defined structure or specific instruction. These systems are efficient for planning quantities or auction bases in supply tenders.
The case for e-procurement in health care
The healthcare sector is poised to benefit greatly from these technologies. They can be used for better identification of the amounts of medicines/medical devices to be purchased and the relevant auction base.
When the need for pharmaceutical and medical products is underestimated, healthcare facilities often find themselves forced to purchase products outside the tender (including through direct procurement) at higher cost. The same problems occur in the case of unsuccessful auctions—a phenomenon that is not infrequent in this sector—when contracting authorities attempt to achieve economies of scale with prices that are so low that suppliers choose not to participate in tenders because the prices set by the administration wouldn’t allow them to turn a profit if they were awarded contracts.[13]
Algorithms for forecasting requirements and contract prices
A forecasting algorithm was the subject of the STEINBOCC Project (Forecasting Tender Needs: Impact Variables and Predictive Data Analytics), conducted by researchers at the Healthcare Data Science Lab at LIUC Università Cattaneo, in cooperation with Egualia (generic, biosimilar, and value added medicines industries).[14] This research developed a forecasting algorithm, accessible via a web interface, that can be adapted for different active ingredients and can be used free of charge by regional contracting authorities and pharmaceutical companies to identify future needs or sizes/amounts for future tenders.
Another predictive algorithm that has been the subject of recent research and is potentially useful in the health sector is the part cost estimator, which aims to identify the optimal price for the auction base, below which the probability of unsuccessful tenders increases.
Some research on the part estimator was carried out by the University of Oviedo, which proposed an ML mechanism for establishing the auction base for tenders for supply of medicines.[15]
Estimation of prices actually paid by administrations is particularly complicated due to the unpredictability of the market and changes to contracts during execution that directly or indirectly affect the cost of supply.
The future scenario
Despite critical present and future issues that wield impact on the digitalization of the procurement lifecycle, the changes introduced by the code in the aggregate are favorable. Mentioning the possible use of automated decision-making systems, including AI systems, sets the stage for design and development of new technologies for public procurement work, though these are prospects for the near future.
The code mentions possibly “using automated procedures” to make decisions, though it does not identify such procedures, nor does it provide any specific indications or examples. Instead, it merely indicates the necessary precautions from a technical point of view to ensure transparency and security when using new technologies.
Still, use of true machine learning algorithms—instead of mere systems for automatic comparison of data and parameters (e.g., algorithms that automatically select the best offer in a tender run under the lowest price criterion)—raises numerous questions about how these algorithms may operate in the context of the “administrative discretion” that governs and guides many choices. It also brings up possible liability deriving from making decisions with automated and AI systems, particularly when their use has direct significant impact on companies that participate in tenders.
One early answer to this question—albeit a partial answer—can be found in Article 30, where human intervention for the purpose of controlling and validating decisions made by automated systems is deemed a necessary part of the algorithm decision-making process. That might lead us to think that this type of human contribution can guarantee that an administration’s discretion in making choices remains established if necessary for the specific activity at hand and if a “person responsible” for decisions made with the contribution of these new technological solutions is identified. In any case, the issue is broad and complex, and the related problems that arise will differ depending on the type of algorithm and where and how it is used in individual tender procedures.
[1] Legislative Decree of March 31, 2023, No.36.
[2] Book I, Part II.
[3] M1C1-70.
[4] M1C1-75.
[5] Art. 19 of the code.
[6] Article 225(2) of the new code provides that the provisions of Articles 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 35, 36, 37(4), 81, 83, 84, 85, 99, the last sentence of 106(3), 115(5), 119(5), and 224(6), as well as Article 4(1)(c) and Article 6(1)(c) of Annex II 4, shall take effect as of January 1, 2024.
[7] This system should allow economic operators and contractors to draft and access documents in digital format, publish and transmit data to the National Public Contracts Database, access all tender documents, submit tenders and the Single European Tender Document, and carry out technical, accounting, and administrative checks on contracts during execution.
[8] To make the transition possible, public administrations had to equip themselves with all the necessary tools, and the ANAC produced a series of resolutions and communications with operational guidelines. Among these, see ANAC Resolution No. 261 of June 20, 2023; ANAC Resolution No 264 of June 20, 2023; ANAC President’s Communiqué of September 19, 2023; ANAC Resolution No. 582 of December 13, 2023; ANAC Resolution No. 601 of December 19, 2023; ANAC Resolution No. 606 of December 19, 2023.
[9] Understanding MIT – ANAC – Resolution No. 582 of 13 December 2023.
[10] Art. 25 co. 2 Procurement Code.
[11] “Procurement: Leli (Fare), flop digitised procurement platforms in healthcare” (link) in Il Sole 24 Ore / Sanità24 of 23 January 2023.
[12] García Rodríguez, Manuel J., et al., “Bidders Recommender for Public Procurement Auctions Using Machine Learning: Data Analysis, Algorithm, and Case Study with Tenders from Spain” (link); García Rodríguez, Manuel J., et al., “Collusion Detection in Public Procurement Auctions with Machine Learning Algorithms” (link); Torres-Berru, Yeferson, and Vivian F. López Batista, “Data Mining to Identify Anomalies in Public Procurement Rating Parameters” (link).
[13] On these topics, see the document titled The Evolution of the Drug and Device Purchasing System (link) developed by the Italian Society of Hospital Pharmacy and Pharmaceutical Services of Health Authorities (SIFO) with the contribution of companies and associations in the sector; see in particular Chapter 9.
[14] “Medicines: An algorithm to manage tenders and avoid shortages” (link) in Il Sole 24 Ore/ Sanità24 of June 7, 2023.
[15] “Public Procurement Announcements in Spain: Regulations, Data Analysis, and Award Price Estimator Using Machine Learning” (link).