BedreFlyt: Improving Patient Flows through Hospital Wards with Digital Twins
Riccardo Sieve, Paul Kobialka, Laura Slaughter, Rudolf Schlatte, Einar Broch Johnsen, Silvia Lizeth Tapia Tarifa · 2025
Read BedreFlyt: Improving Patient Flows through Hospital Wards with Digital Twins by Riccardo Sieve, Paul Kobialka, Laura Slaughter, Rudolf Schlatte, Einar Broch Johnsen, Silvia Lizeth Tapia Tarifa (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-06287.txt.
Introduction
Riccardo Sieve and co-authors describe a digital twin for hospital wards in a 2025 paper. A digital twin is a live model of a real ward that planners can test ideas on. Bed space is tight, and poor plans leave patients waiting while beds sit in the wrong rooms.
Their twin turns each new stream of patients into a set of planning puzzles. A solver then finds bed layouts that fit rules drawn from a shared knowledge base. Planners can test both short-term choices and long-term plans, from daily bed needs to worst-case surges. The authors show the design at work on the task of assigning beds in bays.
Catalog record
| CREATOR | Riccardo Sieve, Paul Kobialka, Laura Slaughter, Rudolf Schlatte, Einar Broch Johnsen, Silvia Lizeth Tapia Tarifa |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-06287
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
BedreFlyt: Improving Patient Flows through Hospital Wards with Digital Twins Riccardo Sieve, Paul Kobialka, Laura Slaughter, Rudolf Schlatte, Einar Broch Johnsen, Silvia Lizeth Tapia Tarifa Abstract Digital twins are emerging as a valuable tool for short-term decision-making as well as for long-term strategic planning across numerous domains, including process industry, energy, space, transport, and healthcare. This paper reports on our ongoing work on designing a digital twin to enhance resource planning, e.g., for the in-patient ward needs in hospitals. By leveraging executable formal models for system exploration, ontologies for knowledge representation and an SMT solver for constraint satisfiability, our approach aims to explore hypothetical “what-if” scenarios to improve strategic planning processes, as well as to solve concrete, short-term decision-making tasks. Our proposed solution uses the executable formal model to turn a stream of arriving patients, that need to be hospitalized, into a stream of optimization problems, e.g., capturing daily inpatient ward needs, that can be solved by SMT techniques. The knowledge base, which formalizes domain knowledge, is used to model the needed configuration in the digital twin, allowing the twin to support both short-term decision-making and long-term strategic planning by generating scenarios spanning average-case as well as worst-case resource needs, depending on the expected treatment of patients, as well as ranging over variations in available resources, e.g., bed distribution in different rooms. We illustrate our digital twin architecture by considering the problem of bed bay allocation in a hospital ward. 1 Introduction Digital Twins (DTs) are virtual information constructs that capture the structure, context, and behavior of the system they are twinning, are dynamically updated with data from the twinned system, have predictive capability, and inform decisions that realize value, according to a recent definition by the National Academy for Science, Engineering and Medicine (NASEM) [ 34 ] . While many applications of DTs so far can be found in engineering disciplines, based on the idea of creating an increasingly accurate “virtual replica” of a physical system to predict behavior by means of sophisticated simulation techniques and a closed feedback loop to a twinned cyber-physical system (e.g., [ 12 ] ), this recent definition is broader. In fact, DTs are today used in a variety of domains outside of cyber-physical systems, including healthcare [ 42 ] , manufacturing [ 6 ] , and transportation [ 8 ] . We believe DTs have significant potential as a tool for the model-driven exploration of so-called “what if” scenarios, moving from the predictive analysis of near-future events to the prescriptive analysis of hypothetical scenarios. Thus, DT technology appears to be useful both for short-term decision-making and long-term strategic planning in various domains. If we consider DTs from a formal methods perspective, DTs differ from standard model-driven techniques by supporting the dynamic update of the model, leveraging live data from the modeled system (often referred to as the physical twin). We may think of the DT as an infrastructure for data-driven formal methods (e.g., [ 26 ] ), in which the stream of data from the twinned system is used to configure a formal model. Similarly, the what-if scenario requested by the user of the DT may determine other aspects of the model as well as the properties to be analyzed. This way, the DT infrastructure may be thought of as a self-adaptive system [ 43 ] for advanced model management, generating the models and determining the analyses to be performed over these models. Our focus here is on DTs for resource management in healthcare, a domain in which resource management is crucial to efficient operations [ 33 ] . The proper handling of resources at a hospital is concerned with how, e.g., trained staff, bed availability in the ward, and necessary rooms and equipment match the needs of the different activities of the hospital, such as the treatment of patients. Efficient resource allocation provides a structured way to better manage workflow and adjust it dynamically, avoiding bottlenecks in operations, thereby allowing for better prioritization and utilization of staff [ 45 ] . Simulations have been successfully used to improve resource allocation in a hospital [ 40 ] ; a DT can connect such simulation models to live data to ensure that the simulation models do not deviate from the resource allocation problem of the hospital. This way, a DT can become the point of contact between static planning and dynamic optimization, allowing a better management of the workflow and making it possible to adjust it dynamically. By configuring the models to explore different scenarios, it is possible to compare different strategies under different assumptions with respect to resources as well as incoming patients. This paper discusses the initial design of BedreFlyt ( \textipa /”be:dr@ fly:t/ ), a DT for resource planning in a hospital ward, with a particular focus on bed bay allocation. The DT combines a knowledge base formalizing knowledge about patient treatments and the ward, an actor-based executable formal model to explore scenarios for streams of incoming patients with associated treatments, and a constraint solver to perform the actual bed allocation. Technically, we combine the executable modeling language ABS [ 18 , 19 ] with the SMT solver Z3 [ 10 ] and knowledge graphs [ 17 ] . The orchestration language SMOL [ 23 , 22 ] is used to combine knowledge graphs with the ABS and Z3 models. SMOL supports querying a knowledge base, which includes the reflection of the runtime state of the SMOL program itself, via SPARQL and SHACL queries. The ABS model transforms the stream of patient data into a stream of constraint problems that capture the bed allocation problem at different points in time. Combined with a description of the ward, these are turned into SMT problems, which we give to Z3. The result is the multi-day planning of bed allocation for patients, that minimizes bed reallocation. We evaluate the design on a realistic patient treatment scenario, based on a historical dataset for a hospital ward at Rikshospitalet, 1 1 1 https://www.oslo-universitetssykehus.no/steder/rikshospitalet/ and further consider how the design scales to larger scenarios up to 2000 patients. Main contributions: a DT design combining formalized knowledge, executable formal models and SMT to explore “what if” scenarios, its implementation, and its application to a hospital ward scenario. Paper overview: Section 2 motivates our work by a hospital ward planning problem, Sect. 3 reviews background and Sect. 4 gives an overview of our DT design. Then Sect. 5 discusses our prototype implementation, Sect. 6 experimental results, and Sect. 7 perspectives on the DT design and future work. 2 Motivating Scenario Resource planning in hospitals is surprisingly complex. Already the problem of allocating beds to patients who are admitted for multi-day stays at a hospital ward is a dynamic scheduling problem concerned with how the patients’ beds are placed in so-called bed bays, i.e. designated slots inside the different rooms of the hospital ward where rolling beds are placed. Additionally, the patients’ bed needs may vary over time, and are subject to numerous constraints; for example, to allocate a bed to a bed bay in the hospital ward one needs to consider whether the patient requires continuous monitoring from staff and the extent to which the patient can share a