Hubbry Logo
Nurse scheduling problemNurse scheduling problemMain
Open search
Nurse scheduling problem
Community hub
Nurse scheduling problem
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Nurse scheduling problem
Nurse scheduling problem
Nurse scheduling problem
View on Wikipedia
from Wikipedia

The nurse scheduling problem (NSP), also called the nurse rostering problem (NRP), is the operations research problem of finding an optimal way to assign nurses to shifts, typically with a set of hard constraints which all valid solutions must follow, and a set of soft constraints which define the relative quality of valid solutions.[1] Solutions to the nurse scheduling problem can be applied to constrained scheduling problems in other fields.[2][3]

While research on computer-assisted employee scheduling goes back to the 1950s,[4] the nurse scheduling problem in its current form was introduced in two parallel publications in 1976.[5][6] It is known to have NP-hard complexity.[1]

General description

[edit]

Conventionally, hospital nursing is shift work that is divided up to provide coverage 24 hours per day, 7 days per week. The hospital has restrictions and requirements for what coverage is needed, and each nurse has their own wishes and restrictions as well. The problem is described as finding a schedule that fulfills the objectives of the hospital and covers all shifts, while respecting as many of the nurses' preferences as possible.

The problem is not unique to nursing. It applies in any other profession or situation where shift coverage must be planned out.

Constraints

[edit]

Creating a schedule means attempting to satisfy certain constraints on how that schedule is laid out. There are two types of constraint: hard constraints, which must be met for the schedule to be valid; and soft constraints, which are desirable but not mandatory.

Depending on the policies of a hospital, the preferences of individual nurses may be treated as either a soft constraint,[7] or as a hard constraint.[8]


Hard constraints may include physical limitations or legal requirements. Some examples of possible hard constraints are:

  • All shifts require nursing coverage.
  • A nurse cannot work more more than one shift at the same time.
  • A nurse cannot work more than 24 hours in a day, or more than 7 days in a week.
  • A nurse must not work more than a legally specified number of days in a row.
  • A nurse must have a legally specified number of rest hours between shifts.
  • Any newly licensed nurse must be paired with an experienced nurse.
  • There must always be one charge nurse on duty.
  • Certain shifts must be covered by nurses with special qualifications.[9]


Soft constraints may be hospital policies or nurse preferences. Some examples of possible soft constraints are:

  • All nurses should work approximately the same number of weekend shifts.
  • All nurses should work approximately the same difficulty in shift assignments.
  • A nurse should not work a day shift and a night shift without a rest day in between.
  • One nurse prefers to have all their work days in a row, and then have all their days off in a row.
  • One nurse prefers to work no more than two consecutive days, with a day off in between.
  • One nurse cannot work Wednesday each week because they have no child care available that day.
  • Two nurses feel they work well together and prefer to be scheduled to work together.


Solutions

[edit]

Solutions to the problem use a variety of techniques, including both mathematically exact solutions[7] and a variety of heuristic solutions using decomposition,[10] parallel computing,[10][11] stochastic optimization,[1] genetic algorithms,[7] colony optimization,[7] simulated annealing,[7] quantum annealing,[12] Tabu search,[7] and coordinate descent.[11][13]

Burke et al. (2004)[14] summarised the state of art of academic research to the nurse rostering problem, including brief introductions of various then published solutions.

See also

[edit]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Nurse scheduling problem

View on Grokipedia
from Grokipedia
The nurse scheduling problem (NSP), also known as the nurse rostering problem (NRP), is a challenge in that involves assigning nurses with varying skills to specific shifts over a defined horizon, such as a week or month, to fulfill staffing demands while adhering to legal, institutional, and individual constraints. This problem balances the need for adequate patient care coverage with factors like shift rotations, rest periods, and workload distribution to ensure operational efficiency and staff well-being. First systematically addressed in the mid-1970s through mathematical programming formulations, the NSP has garnered extensive research attention due to its real-world impact on healthcare delivery, with over 1,000 scholarly articles published on the topic since then. Proven to be NP-hard, the problem's arises from the interplay of numerous constraints and objectives, rendering exact solutions feasible only for small instances while necessitating and approaches for practical scales. Key hard constraints—which must be strictly satisfied—include minimum staffing requirements per shift, maximum consecutive working hours (e.g., no more than two shifts per day), and mandatory rest periods (e.g., at least two off days after a single shift or three after consecutive shifts). Soft constraints, which allow violations at a penalty cost, often encompass nurse preferences such as minimum night or evening shifts per period (e.g., at least one night shift weekly) and total shift limits (e.g., 6–8 per week). The primary objectives of NSP formulations typically aim to minimize operational costs, such as payments, while maximizing nurse satisfaction and equitable workload distribution to reduce burnout and improve retention. Solution methodologies have evolved from early models to advanced techniques like matheuristics, hyper-heuristics, and hybrid AI-driven algorithms, with metaheuristics remaining the most prevalent due to their effectiveness on benchmark datasets like those from the International Nurse Rostering Competition (INRC). Post-COVID-19 trends as of 2025 emphasize AI integration for real-time adaptability, multi-skill considerations, preference elicitation via mobile apps, and acuity-based scheduling to enhance schedule fairness and hospital resilience.

Overview

Definition and Scope

The nurse scheduling problem (NSP) is a challenge that involves assigning nurses to shifts over a specified period to ensure sufficient coverage for care needs, while balancing workload distribution and incorporating staff preferences to the extent feasible. This assignment process must account for the diverse skills and availability of nurses to maintain in healthcare environments like hospitals and clinics. The scope of the NSP varies by institution but generally covers planning horizons from one week to one month, allowing for rosters that align with operational cycles such as periods or seasonal demands. It typically involves teams ranging from small units of 10–30 nurses to larger departments exceeding 100 staff members, depending on the ward size and volume. Common shift types include day (morning/afternoon), night, and weekend rotations, often structured in 8- or 12-hour blocks to provide 24-hour coverage. The NSP is recognized as an NP-hard combinatorial optimization problem because the number of feasible schedules grows exponentially with the number of nurses, shifts, and periods, making exact solutions computationally intractable for realistic instances without approximation methods. This complexity stems from the need to simultaneously satisfy multiple interdependent decisions in a discrete search space. At its core, the problem comprises key elements: a set of nurses with individual attributes, a of shifts across days or weeks, and predefined coverage requirements specifying the minimum number of nurses needed per shift to meet care demands. Constraints related to coverage and regulatory limits on working hours are essential to the formulation, with further details provided in dedicated sections.

Historical Development

In the early 20th century, nurse scheduling in hospitals relied heavily on manual processes, typically managed by head nurses who assigned shifts based on immediate needs, staff availability, and basic preferences, often resulting in significant inefficiencies such as uneven workloads, scheduling errors, and nurse burnout due to and lack of predictability. These practices were time-consuming and prone to subjectivity, exacerbating staffing shortages in growing healthcare systems. The nurse scheduling problem emerged as a formal challenge in the mid-20th century, with initial mathematical models appearing in the , such as the controlled variable staffing approach by Wolfe and Young (1965). By the and into the , the field gained momentum through seminal works like those of Warner and Prawda (1972), who formulated the problem using goal programming to balance coverage requirements and nurse preferences, and Miller et al. (1976), who developed a to minimize deviations from desired staffing levels across shifts. These efforts highlighted the NP-hard nature of the problem and laid the groundwork for optimization techniques in healthcare staffing. During the and , the advent of affordable enabled a transition to computer-assisted tools, with formulations becoming prominent for handling complex constraints like shift coverage and legal regulations. Key developments included goal programming models, as explored by and Ravindran (1981), which utilized techniques including branch-and-bound and were refined in the . The saw the introduction of , a declarative approach that excelled in managing intricate rules and preferences, as demonstrated in early implementations like the HOROPLAN system for generating feasible schedules efficiently. The 2000s brought methods to the forefront, addressing the limitations of exact approaches for real-time, large-scale problems by incorporating search techniques such as and genetic algorithms to approximate optimal rosters while prioritizing nurse satisfaction and equity. These advancements were driven by increasing computational power and the need for flexible systems in diverse settings. The in the 2020s accelerated demand for dynamic modeling in nurse scheduling, as fluctuating patient loads, infection risks, and staff shortages necessitated adaptive algorithms capable of real-time adjustments to maintain coverage without compromising . This period underscored the evolution toward integrated, resilient systems that incorporate uncertainty and rapid response capabilities.

Problem Formulation

Objectives

The primary objectives in the nurse scheduling problem (NSP) revolve around generating rosters that ensure adequate staffing while optimizing and personnel . Hard objectives focus on achieving complete coverage of required shifts to prevent understaffing, which is essential for maintaining care standards and complying with regulatory requirements. These objectives are non-negotiable and form the foundation of any feasible schedule, typically enforced as strict constraints in optimization models. Soft objectives, in contrast, aim to improve the quality of the schedule beyond basic coverage, allowing for trade-offs to enhance overall performance. Key examples include minimizing total costs, such as payments or premium shifts, to control expenses; maximizing nurse satisfaction by fulfilling preferences for specific days off, shift types, or consecutive free days; and balancing workload to promote equity among staff, for instance, by distributing night shifts or total hours evenly. These goals address both economic pressures on healthcare providers and human factors like job retention and morale. The NSP often involves , where trade-offs must be navigated, such as reducing costs at the potential expense of fairness or satisfaction. Seminal work introduced goal programming to handle these conflicts by prioritizing or weighting competing aims, allowing decision-makers to balance efficiency against staff equity. Mathematically, objectives are commonly aggregated into a single function using weighted linear combinations, where the goal is to minimize the total penalty or cost. A representative is: min⁡∑i(ci⋅oi+pi⋅vi+f⋅σh)\min \sum_{i} \left( c_i \cdot o_i + p_i \cdot v_i + f \cdot \sigma_h \right) Here, ci⋅oic_i \cdot o_i represents costs for nurse ii, pi⋅vip_i \cdot v_i denotes penalties for unmet preferences (e.g., requested days off), σh\sigma_h is the standard deviation of total hours worked across nurses as a , and coefficients weight their relative importance. This structure enables optimization while respecting coverage requirements. Evaluation of schedules typically relies on metrics like total operational cost, which quantifies financial impact, and fairness indices such as the variance in assigned hours or shifts, which assess workload equity (lower variance indicates better balance). These metrics provide quantifiable benchmarks for comparing solution quality in practice.

Constraints

The nurse scheduling problem incorporates a variety of constraints that ensure feasible and equitable rosters, broadly categorized into hard constraints, which must be strictly satisfied to produce a valid schedule, and soft constraints, which represent desirable but flexible conditions often incorporated via penalties. Hard constraints typically include coverage requirements, legal limitations on working hours, and skill matching to maintain and compliance. Coverage constraints mandate that the minimum number of nurses assigned to each shift meets the demand for patient care, such as ensuring at least three registered nurses are available during a night shift in a general ward. Legal limits enforce regulations like the European Union's Working Time Directive, which restricts average weekly working hours to 48 (including overtime) and requires at least 11 consecutive hours of rest per day, or U.S. state-specific laws, such as California's prohibition on shifts exceeding 12 hours except in emergencies. Skill matching constraints require assigning nurses only to shifts or units where they possess the necessary qualifications, for instance, directing nurses with intensive care to critical care areas while prohibiting unqualified staff from specialized pediatric duties. Soft constraints address nurse preferences, fairness in workload distribution, and institutional policies to enhance satisfaction and retention without invalidating the schedule. Examples include accommodating requests for specific days off or preferred shift types, such as avoiding early mornings for nurses with childcare responsibilities; promoting fairness by limiting excessive weekend assignments, like no more than two consecutive working weekends per nurse; and adhering to policies like balanced holiday coverage to distribute undesirable shifts evenly. In formal mathematical representations, the problem is often modeled using , where xi,jx_{i,j} is a binary variable indicating whether nurse ii is assigned to shift jj. Coverage is enforced by the inequality ∑ixi,j≥dj\sum_{i} x_{i,j} \geq d_j for each shift jj with demand djd_j, while other hard constraints appear as similar linear inequalities or bounds on variables. Soft constraints are handled by incorporating violations into the objective function as penalties, such as adding a term P∑kvkP \sum_{k} v_k where vkv_k measures the degree of violation for soft constraint kk and PP is a penalty weight, allowing trade-offs during optimization.

Solution Methods

Exact Approaches

Exact approaches to the nurse scheduling problem, also referred to as the nurse rostering problem, employ techniques that guarantee globally optimal solutions by exhaustively exploring the solution space while respecting all hard constraints. These methods are computationally demanding due to the problem's combinatorial nature but are effective for instances with up to a few dozen nurses and planning horizons of a few weeks. They typically outperform heuristics in solution quality for solvable sizes, providing verifiable optimality certificates useful for validation or bounding larger approximations. Integer programming (IP) formulates the problem as a 0-1 integer linear program, where binary decision variables indicate whether a specific nurse is assigned to a particular shift on a given day. Constraints model requirements such as shift coverage, maximum consecutive shifts, and rest periods, while the objective minimizes penalty costs for soft constraints like preference violations. Solvers like IBM ILOG CPLEX apply branch-and-bound algorithms to prune infeasible branches or cutting planes to tighten relaxations, ensuring exact solutions. For example, a pure IP model implemented in CPLEX handles binary variables for shift assignments and changeovers, solving Danish hospital instances with coverage and overlap constraints. Constraint programming (CP) represents variables as domains of possible shift assignments for each nurse-day pair, with propagators enforcing constraints through domain reduction. Key features include all-different constraints to ensure unique shift allocations per day across nurses and global constraints for patterns like minimum rest times between shifts. CP advantages lie in efficiently managing logical and temporal constraints via constraint propagation, often outperforming IP on subproblems involving feasibility checks. Implementations using IBM ILOG CP Optimizer model nurse rosters as constraint satisfaction problems, generating feasible weekly sub-schedules before optimization. Mixed-integer linear programming (MILP) builds on IP for multi-objective nurse scheduling by incorporating lexicographic or weighted sums to balance goals like equitable workload distribution and preference satisfaction. These extensions use continuous variables for aggregated measures alongside binaries for assignments, solved via standard MILP solvers such as CPLEX or Gurobi with branch-and-cut procedures. A flexible MILP system for real-world rosters integrates multi-stage planning, where pre-assignments precede full optimization, handling soft constraints through penalty terms. The nurse scheduling problem is NP-hard, implying exponential worst-case for exact solutions, even for decision versions checking feasibility. Benchmarks from the International Nurse Rostering Competitions (INRC-I and INRC-II) demonstrate this, with datasets featuring 20–60 nurses over 28 days requiring advanced techniques like branch-and-price hybrids for tractability. Exact methods thus provide lower and upper bounds that inform strategies for intractable scales.

Heuristic and Metaheuristic Approaches

Heuristic and approaches provide scalable solutions to the nurse scheduling problem (NSP) by generating high-quality rosters efficiently for large instances where exact methods are computationally prohibitive. These methods start with an initial feasible schedule and iteratively refine it through neighborhood explorations, balancing coverage constraints, nurse preferences, and workload fairness without guaranteeing global optimality. Local search form the foundation, while metaheuristics enhance exploration to avoid local optima, often achieving solutions within 95-99% of the optimum in minutes for schedules involving over 100 nurses. Local search heuristics, such as , iteratively improve an initial roster by evaluating and accepting neighboring solutions that reduce the objective function, typically through shift swaps or assignments that repair violations. This deterministic approach excels in refining feasible schedules but risks stagnation in local optima, as demonstrated in competitive solutions for the International Nurse Rostering Competition (INRC-I) where was combined with decomposition techniques to handle complex constraints. extends hill climbing by incorporating probabilistic acceptance of worse solutions early on, inspired by metallurgical annealing, to escape local traps and converge to better global approximations. Applied to real-world Italian hospital instances, with large neighborhoods produced competitive rosters satisfying hard constraints like minimum staffing while minimizing soft violations such as . Metaheuristics build on local search by introducing mechanisms for broader exploration. Genetic algorithms (GAs) represent schedules as chromosomes—arrays encoding nurse-shift assignments—and evolve populations through crossover, , and selection to optimize multi-objective fitness functions incorporating preferences and equity. An indirect GA encoding, using priority-based permutations to generate rosters, effectively solved hospital problems by avoiding infeasible intermediates and outperforming manual scheduling in preference satisfaction. enhances local search by maintaining a of recent moves to prevent , allowing diversification through strategic oscillation between feasible and infeasible regions; this yielded near-expert quality schedules in early applications, closely matching human-generated rosters for coverage and fairness. Variable neighborhood search (VNS) systematically varies neighborhood structures—from simple swaps to complex reassignments—to systematically explore the solution space, proving robust for benchmark instances. A randomized VNS variant achieved state-of-the-art results on INRC-I datasets by greedily repairing initial solutions and iteratively applying diverse operators, often surpassing other metaheuristics in solution quality for multi-skill environments. (PSO), adapted for discrete NSP, models nurses as particles updating positions toward personal and global best schedules via velocity adjustments and mutations, effectively balancing workload in settings. A guided PSO for Malaysian public hospitals minimized understaffing while respecting rest constraints, demonstrating convergence in under 100 iterations for 30-nurse instances. Hybrid approaches, including matheuristics, integrate heuristics with exact methods like (IP) for validation or . For instance, a matheuristic using IP to generate initial partial rosters followed by VNS refinement solved extended NSP variants with skills and last-minute changes, achieving 98% optimality gaps on benchmarks. These hybrids leverage the strengths of both paradigms, using exact solvers for subproblems while heuristics scale to practical deployments. Performance evaluations on INRC-II benchmarks highlight their efficacy, with winners producing rosters 5-10% better than baselines in multi-stage planning, often validated against IP optima for smaller cases.

Applications and Extensions

Real-World Implementations

In healthcare settings, commercial software solutions for nurse scheduling often combine for exact optimization with methods to handle complex, real-time constraints, enabling efficient . For instance, Clinical Scheduling (formerly Kronos Workforce Scheduler) uses acuity-based algorithms to track hours, maintain staffing ratios, and generate balanced schedules that minimize overtime while adhering to labor regulations. Similarly, ShiftWizard by HealthStream employs cloud-based s to automate shift assignments, integrate staff preferences, and forecast demand, thereby reducing administrative burden and improving compliance in hospitals and clinics. These tools have been adopted by thousands of healthcare organizations to streamline operations without requiring extensive manual intervention. Case studies illustrate the practical deployment of these systems. In a 2013 hospital implementation in , a genetic algorithm-based scheduler reduced expenses by 12% and the number of nurses by 13% while achieving a fairer distribution of pay, leading to more predictable workloads. These examples highlight how algorithmic approaches translate to tangible operational improvements in diverse regulatory environments. Integration with electronic health records (EHR) systems allows nurse scheduling tools to adjust dynamically to patient demand, pulling on acuity levels and fluctuations for proactive staffing. For example, Epic's scheduling modules, such as , within the Epic EHR platform, synchronize staff assignments with patient flow, enabling automatic updates to rosters based on admission forecasts and reducing under- or over-staffing. This linkage supports enforcement of constraints like legal shift limits by incorporating clinical data directly into optimization models. Deployments of nurse scheduling solutions have yielded measurable impacts on healthcare operations. Automated systems have contributed to lower nurse turnover rates in facilities emphasizing flexible scheduling, as self-scheduling options enhance and work-life balance. Cost savings are evident in reductions of up to 50% in agency staffing expenses, achieved through better internal shift coverage that minimizes reliance on external contractors. Additionally, HIPAA-compliant features in tools like Connecteam ensure secure handling of during roster creation and sharing, promoting regulatory adherence without compromising efficiency. Post-pandemic adaptations from 2020 onward have focused on flexible rostering frameworks to address fluctuating demands. Hospitals developed strategies that scaled staffing by reallocating non-critical care nurses to high-acuity units and enhancing coverage through redeployment and new roles. These models, informed by experiences, prioritized nurse safety through enforced rest periods. As of 2025, ongoing integrations of AI-driven predictive tools continue to support dynamic adjustments in response to evolving healthcare needs.

Adaptations in Other Fields

The nurse scheduling problem (NSP) framework, characterized by constraints on coverage, rest periods, and requirements, has been adapted to personnel rostering in various non-healthcare domains, where similar models address workforce allocation under temporal and regulatory pressures. In , NSP adaptations focus on pilot and crew rostering, incorporating flight coverage, mandatory intervals, and qualification-based assignments to ensure operational safety and compliance with aviation regulations. These problems often decompose into pairing generation (assigning crew to flight legs) and personalized rostering stages, using models akin to NSP formulations. For instance, techniques solve large-scale instances involving irregular flight schedules and constraints, achieving near-optimal solutions for major airlines. Transportation applications, particularly bus driver shift scheduling, adapt NSP by emphasizing temporal coverage of predefined routes, driver rest requirements, and relief point constraints to minimize idle time and overtime. Methods such as set partitioning and , drawn from NSP literature, generate feasible rosters that balance workload fairness and service reliability. A combined bus-driver scheduling approach, for example, integrates vehicle and personnel assignment using network flow models similar to those in NSP, reducing total costs by up to 10% in real transit networks. In scheduling, NSP principles underpin assignments, where multi-objective models ensure fair distribution of games across officials, respecting availability, travel distances, and expertise levels to avoid conflicts and . These adaptations prioritize equity in , much like nurse preferences in NSP, and have been applied to leagues such as and , yielding balanced rosters that enhance officiating quality. Manufacturing shift planning for assembly lines extends NSP by incorporating skill-specific constraints and production demands, assigning workers to shifts that maintain line coverage while adhering to labor laws on rest and . Hybrid approaches, inspired by NSP heuristics, optimize these schedules to minimize disruptions and costs in high-volume environments. A key adaptation in retail staffing involves modifying NSP objectives from coverage maximization to profit optimization, where schedules align employee availability with periods to boost sales while controlling wage expenses. variants forecast customer traffic and assign shifts accordingly, improving revenue by 5-15% in store chains through dynamic adjustments. Airline crew scheduling software, such as systems derived from NSP frameworks, employs hybrid exact-heuristic methods to handle rosters for over 1,000 employees, integrating real-time disruptions like with constraints on and pairings for efficient global operations.

Challenges and Future Directions

Current Limitations

One major limitation in solving the nurse scheduling problem lies in , particularly with exact methods such as integer or branch-and-price algorithms, which become computationally infeasible for large-scale instances typical of major hospitals. For example, these approaches often require excessive time—sometimes hours or days—to solve problems involving more than 120 nurses over multiple weeks, rendering them impractical for real-time or daily scheduling needs in facilities with hundreds of staff members. methods can address speed but introduce approximation errors that may violate constraints in complex scenarios. Another significant challenge is handling in real-world environments, where static models fail to account for variable demand such as fluctuating patient admissions, unexpected absences, or changes in acuity levels. Traditional deterministic formulations assume fixed inputs, leading to suboptimal or infeasible rosters when disruptions occur, as they lack robust elements to incorporate probabilistic forecasts of these variables. This gap is particularly acute in dynamic settings like emergency departments, where demand variability necessitates reactive adjustments that strain resources. Incorporating human factors presents further difficulties, as models struggle to integrate subjective nurse preferences—such as desired shift patterns for work-life balance—alongside like communication needs or impacts, often resulting in biased or incomplete representations. Quantitative objectives in optimization frameworks typically prioritize coverage over qualitative aspects, leading to schedules that overlook individual and foster dissatisfaction or burnout. For instance, without nuanced modeling of interpersonal preferences, algorithms may inadvertently create rosters that disrupt team cohesion, exacerbating turnover in affected units. Data privacy issues arise prominently in AI-driven approaches, which rely on personal nurse data like preferences, health records, and performance metrics, complicating compliance with regulations such as the General Data Protection Regulation (GDPR) in the . These models process sensitive information to generate personalized schedules, but inadequate anonymization or mechanisms breaches, with potential fines up to 4% of annual turnover for non-compliance. Balancing algorithmic accuracy with privacy-preserving techniques, like , remains underdeveloped in nurse scheduling applications. Finally, equity gaps persist as many models overlook diversity factors, such as , cultural needs, or levels, often producing unfair rosters that disproportionately burden junior or underrepresented nurses. Optimization frameworks frequently default to uniform rules without accounting for these variables, leading to biased allocations where, for example, senior staff receive preferential shifts, contributing to perceptions of inequity and higher attrition among diverse groups. This oversight can widen existing disparities in the predominantly female , where -specific needs like maternity leave integration are inadequately addressed. Recent advancements in and are transforming the nurse scheduling problem by enabling dynamic and predictive capabilities. Reinforcement learning approaches, such as the attention-guided NurseSchedRL framework, have been developed to handle dynamic rescheduling under , including sudden nurse absences, by learning optimal strategies that generate feasible rosters in real-time environments. Similarly, neural networks trained on (EHR) data are being utilized to predict staffing demands and workload fluctuations, allowing for proactive adjustments; for example, models derived from EHRs predict nursing workload to support data-driven staffing decisions. These 2023–2025 studies demonstrate improved adaptability, with RL models outperforming traditional methods in uncertain scenarios by up to 20% in schedule feasibility. Mobile and app-based platforms are gaining traction for facilitating real-time nurse preferences and automated rostering. In 2025 pilots, applications designed for objective scheduling enable nurses to submit shift preferences via mobile interfaces, automating roster generation to enhance fairness and efficiency; these tools have been shown to save significant planning time while boosting nurse satisfaction in hospital settings. Pandemic-informed models post-2022 emphasize seniority-based assignments and workload balancing to cope with infectious disease surges. has introduced mathematical formulations that categorize nurses by experience levels and permit flexible consecutive shifts, ensuring balanced distribution during high-demand periods like outbreaks, as validated in real case studies from healthcare facilities. A growing focus is emerging in nurse scheduling, particularly green approaches that minimize travel distances and emissions in home contexts. models prioritize low-emission routing for staff assignments, reducing CO₂ output while maintaining workload equity, with potential extensions to multi-site networks. Future prospects include deeper integration with (IoT) technologies for real-time staffing decisions, such as wearables that monitor nurse fatigue to trigger schedule adjustments and prevent burnout. Benchmarks from ongoing competitions, including the International Nurse Rostering Competition (INRC) series as of 2025, continue to drive innovation by evaluating these advancements against standardized instances.

References

  1. https://www.researchgate.net/publication/360821639_A_Survey_of_the_Nurse_Rostering_Solution_Methodologies_The_State-of-the-Art_and_Emerging_Trends
  2. https://www.sciencedirect.com/science/article/pii/S111001681730282X
  3. https://www.researchgate.net/publication/220366435_The_State_of_the_Art_of_Nurse_Rostering
  4. https://www.graham-kendall.com/papers/ngssak2022a.pdf
  5. https://www.healthstream.com/resources/2025-trends-in-nurse-scheduling
  6. https://pmc.ncbi.nlm.nih.gov/articles/PMC12406402/
  7. https://www.nature.com/articles/s41598-019-49172-3
  8. https://www.sciencedirect.com/science/article/pii/S037722170300095X
  9. https://pmc.ncbi.nlm.nih.gov/articles/PMC8035616/
  10. https://www.growingscience.com/msl/Vol2/msl_2012_143.pdf
  11. https://link.springer.com/article/10.1023/B:JOSH.0000046076.75950.0b
  12. https://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=3279&context=lkcsb_research
  13. https://pubmed.ncbi.nlm.nih.gov/14346007/
  14. https://pubsonline.informs.org/doi/10.1287/opre.24.5.857
  15. https://pubmed.ncbi.nlm.nih.gov/7662877/
  16. https://www.tandfonline.com/doi/full/10.1080/08839510600903841
  17. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0275890
  18. https://www.tandfonline.com/doi/full/10.2147/RMHP.S302156
  19. https://d-nb.info/1202043291/34
  20. https://www.tandfonline.com/doi/abs/10.1080/05695558108974536
  21. https://www.schedulingbenchmarks.org/papers/ANROM.pdf
  22. https://www.stern.nyu.edu/om/faculty/pinedo/book2/downloads/Nottingham/11_nurse%20rostering.pdf
  23. https://dol.ny.gov/mandatory-overtime-nurses
  24. https://nursejournal.org/resources/mandatory-overtime-for-nurses/
  25. https://doi.org/10.1016/0038-0121(96)00010-9
  26. https://data.math.au.dk/publications/phd/2013/math-phd-2013-jb.pdf
  27. https://pmc.ncbi.nlm.nih.gov/articles/PMC11675476/
  28. https://people.cs.nott.ac.uk/pszrq/files/AI08.pdf
  29. https://orbit.dtu.dk/files/256583884/JOSH_D_19_00193_accepted.pdf
  30. https://scispace.com/pdf/a-survey-of-the-nurse-rostering-solution-methodologies-the-1bqrn295.pdf
  31. https://www.sciencedirect.com/science/article/pii/S0167637723001177
  32. https://www.sciencedirect.com/science/article/pii/S2211692323000024
  33. https://www.sciencedirect.com/science/article/pii/S0305054803000340
  34. https://www.sciencedirect.com/science/article/pii/S0377221797002816
  35. https://www.sciencedirect.com/science/article/abs/pii/S0038012122000854
  36. https://www.patatconference.org/patat2018/files/proceedings/paper46.pdf
  37. https://www.ukg.com/products/ukg-clinical-scheduling-extensions
  38. https://www.softwareadvice.com/hr/shiftwizard-profile/
  39. https://marketplace.workday.com/en-US/apps/505872/shiftwizard-by-healthstream/overview
  40. https://onlinelibrary.wiley.com/doi/10.1155/2014/246543
  41. https://www.uptech.team/blog/healthcare-scheduling-software-systems
  42. https://sparkco.ai/blog/nurse-shift-scheduling-software-streamline-staffing-for-snfs
  43. https://connecteam.com/top-nurse-scheduling-software-solutions/
  44. https://pmc.ncbi.nlm.nih.gov/articles/PMC12207134/
  45. https://www.nursingworld.org/globalassets/covid19/final-ana-document-csc-staffing-guidance-4.22.21.pdf
  46. https://www.healthstream.com/resources/nurse-scheduling-software-shiftwizard
  47. https://doi.org/10.1016/S0377-2217(03)00095-X
  48. https://www.schedulingbenchmarks.org/papers/new_approaches_to_nurse_rostering_benchmark_instances.pdf
  49. https://www.jise.ir/article_176581_151e124569a063b25f2d106a0fc6ba63.pdf
  50. https://onlinelibrary.wiley.com/doi/10.1155/2022/3293806
  51. https://arxiv.org/html/2505.22124v1
  52. https://www.researchgate.net/publication/392167498_A_Nurse_Staffing_and_Scheduling_Problem_with_Bounded_Flexibility_and_Demand_Uncertainty
  53. https://cosmos.uta.edu/wp-content/uploads/2020/04/COSMOS-07-04.pdf
  54. https://pmc.ncbi.nlm.nih.gov/articles/PMC12157959/
  55. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0256300
  56. https://www.medrxiv.org/content/10.1101/2025.02.18.25322227v1.full-text
  57. https://bmcnurs.biomedcentral.com/articles/10.1186/s12912-025-03808-0
  58. https://pmc.ncbi.nlm.nih.gov/articles/PMC11675209/
  59. https://emergingrnleader.com/ensuring-fairness-to-nursing-staff/
  60. https://bmcnurs.biomedcentral.com/counter/pdf/10.1186/s12912-025-03808-0.pdf
  61. https://www.icn.ch/sites/default/files/2024-01/PS_Gender%2520equity%2520in%2520the%2520nursing%2520and%2520healthcare%2520workforce%2520%2520FINAL%252012.2023_EN.pdf
  62. https://arxiv.org/pdf/2509.18125
  63. https://www.jmir.org/2025/1/e66667
  64. https://www.researchgate.net/publication/393352784_A_Dynamic_Nurse_Scheduling_Approach_Using_Reinforcement_Learning_to_Address_Sudden_Absences_of_an_Unknown_Nurse
  65. https://bmcnurs.biomedcentral.com/articles/10.1186/s12912-025-03370-9
  66. https://www.researchgate.net/publication/364454662_New_nurse_scheduling_model_considering_nurses%27_seniority_and_the_possibility_of_consecutive_shifts_under_COVID-19_A_real_case_study
  67. https://www.sciencedirect.com/science/article/pii/S221471602400006X
  68. https://nrpcompetition.kuleuven-kulak.be/
Add your contribution
Related Hubs
User Avatar
No comments yet.