{"id":5457,"date":"2026-11-06T17:02:12","date_gmt":"2026-11-06T16:02:12","guid":{"rendered":"https:\/\/www.dlvsystem.it\/dlvsite\/it\/?post_type=avada_portfolio&p=5457"},"modified":"2026-11-06T17:02:12","modified_gmt":"2026-11-06T16:02:12","slug":"dlv-hero-project-hospital-employee-roster-optimization","status":"publish","type":"avada_portfolio","link":"https:\/\/www.dlvsystem.it\/dlvsite\/portfolio-items\/dlv-hero-project-hospital-employee-roster-optimization\/","title":{"rendered":"\u201cDLV-HERO\u201d Project: Hospital Employee Roster Optimization"},"content":{"rendered":"
<\/span><\/div><\/div><\/div>
<\/span><\/div><\/div><\/div>
<\/div>
\n

Co-funded by the European Union<\/b><\/p>\n

Programme: PR Calabria ERDF ESF+ 2021-2027<\/b><\/p>\n

Measure: Priority 1 STEP \u00b7 Action 1.6.1 \u2014 Support for investment in digital technologies and deep-tech innovation \u00b7 Call \u201cSTEP Calabria \u2013 Aid for critical technologies\u201d<\/b><\/p>\n

CUP: J25H25000760005<\/b><\/p>\n

COR: 26299099<\/b><\/p>\n

Grant: Calabria Region Decree no. 16871 of 6 October 2026<\/b><\/p>\n

Period: January 2026 \u2014 December 2027<\/b><\/p>\n

The context<\/b><\/p>\n

Planning hospital rosters means juggling hundreds of constraints: employment contracts, mandatory rest periods, the skills each ward and operating theatre requires, staff absences and availability. In many facilities this is still done by hand, or with tools that produce a plan without explaining how they got there. The result is rosters that are hard to check, workloads that are unevenly spread and services that could be offered but are left uncovered.<\/p>\n

The solution<\/b><\/p>\n

DLV-HERO (Hospital Employee Roster Optimization)<\/b> is a platform for the automatic scheduling of healthcare staff, delivered as a cloud service:<\/p>\n

\u2022 a neuro-symbolic AI engine that builds the rosters;<\/p>\n

\u2022 a web application for planners and a mobile app for staff;<\/p>\n

\u2022 integration with hospital information systems and booking centres through the HL7\/FHIR standard, so that optimised rosters turn into new slots patients can book.<\/p>\n

DLV-HERO<\/b> is an organisational support tool: it makes no clinical decisions and does not affect patient care.<\/p>\n

How it reasons<\/b><\/p>\n

At its heart is DLV, our reasoning engine based on Answer Set Programming. The rules (collective agreements, rest periods, skills, minimum staffing) are written declaratively, and the engine only builds plans that satisfy all of them: hard constraints are guaranteed by construction, not \u201cas far as possible\u201d. Every shift can be traced back to the rules that produced it. The result is explainable scheduling, auditable and fairer for staff.<\/p>\n

Alongside logical reasoning, machine learning models learn from historical data, such as absence peaks or seasonal demand, and steer the engine towards better solutions in less time, even for facilities with thousands of employees.<\/p>\n

The pilot<\/b><\/p>\n

The project takes DLV\u2019s healthcare technology from laboratory prototype (TRL 4) to a system demonstrated in a real operating environment (TRL 7), with a pilot in two healthcare facilities in Calabria. The goals:<\/p>\n

\u2022 increase coverage of specialist shifts by 10%;<\/p>\n

\u2022 reduce waiting times in the pilot wards by up to 20%.<\/p>\n

The project is also strengthening DLVSystem\u2019s research team with new roles.<\/p>\n

Funding<\/b><\/p>\n

DLV-HERO<\/b> was selected under the PR Calabria ERDF ESF+ 2021-2027 programme (Priority 1 STEP, Action 1.6.1) and is carried out with resources from the European Regional Development Fund (ERDF), the Italian State and the Calabria Region.<\/p>\n<\/div><\/div><\/div>

<\/div><\/div><\/div><\/div><\/div><\/div>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":4,"featured_media":5459,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","format":"standard","meta":{"footnotes":""},"portfolio_category":[],"portfolio_skills":[],"portfolio_tags":[210,206,212,204,200,214,208],"class_list":["post-5457","avada_portfolio","type-avada_portfolio","status-publish","format-standard","has-post-thumbnail","hentry","portfolio_tags-artificial-intelligence","portfolio_tags-data-analysis","portfolio_tags-industry-4-0","portfolio_tags-machine-learning","portfolio_tags-predictive-maintenance","portfolio_tags-rd","portfolio_tags-system-integration"],"_links":{"self":[{"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/avada_portfolio\/5457","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/avada_portfolio"}],"about":[{"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/types\/avada_portfolio"}],"author":[{"embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/comments?post=5457"}],"version-history":[{"count":3,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/avada_portfolio\/5457\/revisions"}],"predecessor-version":[{"id":5461,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/avada_portfolio\/5457\/revisions\/5461"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/media\/5459"}],"wp:attachment":[{"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/media?parent=5457"}],"wp:term":[{"taxonomy":"portfolio_category","embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/portfolio_category?post=5457"},{"taxonomy":"portfolio_skills","embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/portfolio_skills?post=5457"},{"taxonomy":"portfolio_tags","embeddable":true,"href":"https:\/\/www.dlvsystem.it\/dlvsite\/wp-json\/wp\/v2\/portfolio_tags?post=5457"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}