News

DRONA: Transforming IT Projects with Artificial Intelligence

  • Interdisciplinary Centre for Security, Reliability and Trust (SnT)
    07 octobre 2021
  • Catégorie
    Recherche

Digitalisation is becoming increasingly part of our daily lives, and has been an ongoing process for many decades now. A consequence of this is that, when it comes to financial services, if a business wants to launch a new product or implement new regulations, their success will often lie within their own IT processes. This situation is easier said than done, since upgrading existing systems – or creating new ones entirely – is a notoriously complex process. In 2018, SnT partnered with Clearstream and escent to develop a new tool that would aim to slash error rates and cut production costs, all the while boosting service quality.

The Challenge

The joint research project between Clearstream, the Luxembourg-based, global post-trading financial service provider, and escent, a business analyst consulting firm, set out to solve industry-wide problems. One scenario they wanted to explore involved how artificial intelligence (AI) and natural language processing (NLP) could support requirements analysis for IT projects in the finance industry. In these projects, requirements stipulate exactly what stakeholders wish to change, the user requirements and what the capabilities of the system need to be.

Ordinarily, once these requirements are established, it is then down to a team of technicians and business analysts to make the changes. However, no matter how experienced the team is, human error is inevitable. All too often, the outcome can be sub-optimal, resulting in disappointment and frustration amongst those involved. This project aims to make requirements analysis more efficient and effective, by providing analysts with semi-automated assistance for requirements engineering tasks, such as requirements elicitation and specification. Speaking about this industry problem, SnT’s Prof. Lionel Briand, principal investigator of the project and head of the Software Verification and Validation (SVV) research group said, “Nearly 50% of budget overruns in IT projects are caused by inadequate requirements that ripple through system design and deployment. SnT has brought together a strong team of scientists and engineers to tackle this critical problem.”

The Necessary Requirements

The role of a business analyst involves understanding the needs of clients and internal users, as well as the technological developments required to satisfy those needs. They have to handle the changes to IT systems while addressing stakeholder objectives. But elicitation and management of requirements, while taking care of the minutiae, is hard work. Creating and evolving IT systems is often a highly intricate endeavour that requires a high level of expertise and attention to detail. For even the most skilled and experienced business analysts, this is a challenge. Elene Pitskhelauri, business analyst at Clearstream, said, “We were looking for an assistance that could increase the quality of the analysis and project efficiency.” Their team were looking for a digital tool that would comply with the existing methodology that Clearstream’s busines analysts in Investment Fund Services IT are utilising, and provide support in IT projects.

Put simply, in the same way that word processing packages have templates, as well as spelling or grammar checks, Elene and her colleagues wanted a system that would verify their IT requirements would be written correctly, and based on solid foundations – as well as being internally coherent and correct. They recognised how AI techniques could address this challenge and, more specifically, understood that the technology could improve the efficiency of the IT projects and time-to-market.

An Intelligent Digital Assistant

Once the partnership between SnT, Clearstream and escent had formed, work on creating the technology – DRONA – began. On SnT’s side, the project saw the collaboration between a strong team of scientists. These include Seung Yeob Shin, Mauricio Alferez and Damiano Torre, as well as Angelo Rizzi, a research engineer, and Alvaro Veizaga, a Ph.D. student.

Together they created DRONA, a digital assistant that guides and semi-automates critical IT analysis activities. More definitively, it provides semi-automated support for the definition of system requirements, their verification and the generation of documents. It also provides acceptance criteria that the system is verified against to check its compliancy. This work is all about creating complete, adequate and consistent system requirements, as these are the building blocks of IT projects. The requirements specify what future users can expect from a system, and therefore what needs to be developed.

Relying on AI, as well as automated validation and verification assistance, DRONA ensures the consistency and completeness of requirements to a significant degree. It also ensures that the right functionality is being developed and properly validated, to achieve a successful deployment.

Cost and Time Savings

In taking the time to develop a software that is tailored to Clearstream’s specific needs, the result is an intelligent tool that will help businesses develop higher quality, and more successful information systems. “Cost and time-to-market are slashed, and system quality and customer satisfaction are boosted,” said Thomas Henin, Investment Fund Services IT business analyst lead at Clearstream. “This is because DRONA gives analysts the tailored support they need to build solid foundations for each project,” he continued.

With this being the final year of the project, DRONA will continue to be refined to make it more robust, functional and user-friendly. Members of Clearstream’s team are currently being trained on how to use the software. With this successful proof of concept, DRONA could fill a gap in the market of the financial domain, where thoroughly executed IT projects play a critical role.