Description:FIELD OF DISCLOSURE
[0001] The present disclosure relates generally relates to the field of healthcare informatics and hospital operations management. More specifically, it pertains to a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting.
BACKGROUND OF THE DISCLOSURE
[0002] Efficient hospital management has long been a cornerstone of public health systems across the world. Hospitals, being the primary nodes of healthcare delivery, face significant challenges in managing patient inflow, medical staff availability, and overall resource allocation. Among the many operational difficulties, one of the most critical issues is the unpredictability of patient arrival patterns and the subsequent impact on queue lengths. Unanticipated surges in patient numbers often result in overcrowding, long waiting times, and significant strain on hospital staff and infrastructure. Conversely, underestimation of demand may lead to underutilization of resources, inefficient allocation of staff, and patient dissatisfaction. As a result, forecasting hospital queues and patient arrivals has become an area of intense research, with applications spanning emergency departments, outpatient services, diagnostic facilities, and specialized treatment units.
[0003] Historically, the concept of hospital queues emerged from the broader discipline of queuing theory. Queuing theory was initially applied to industrial and telecommunication systems but later found relevance in healthcare, where waiting times and patient flows became pressing operational concerns. Early models of hospital queues typically relied on deterministic assumptions, such as constant arrival rates or fixed service times. These simplified models provided some theoretical insights but fell short in capturing the stochastic, seasonal, and highly dynamic nature of patient arrivals in real-world hospital settings. The irregularity of human illness, varying levels of emergencies, external factors such as weather or epidemics, and socio-economic determinants all contribute to the inherent complexity of healthcare demand forecasting.
[0004] As healthcare systems expanded and patient populations grew, researchers began exploring time series models for more realistic prediction of hospital queues. Classical statistical methods such as autoregressive integrated moving average (ARIMA) and seasonal ARIMA (SARIMA) became widely adopted due to their ability to model temporal dependencies in patient arrival data. These models provided significant improvements over purely theoretical queuing models by accounting for historical patterns and seasonality. For example, studies demonstrated that hospital emergency departments often experience distinct weekly or monthly patterns, with weekends and holidays showing marked differences in patient flow. However, despite their usefulness, ARIMA-type models struggled with handling multiple seasonalities and irregular patterns that are characteristic of hospital data. This limitation became particularly evident in large urban hospitals, where patient arrivals are influenced by numerous overlapping factors, including traffic conditions, outbreak trends, demographic shifts, and policy changes.
[0005] The increasing complexity of healthcare systems prompted the exploration of advanced statistical models capable of capturing multiple levels of seasonality and irregularities. Among these, the Trigonometric Box-Cox ARIMA Trend Seasonal (TBATS) model gained prominence for its suitability in modeling time series with complex seasonal structures. TBATS introduced features such as Fourier terms for seasonality representation, Box-Cox transformations for variance stabilization, and state-space frameworks for parameter estimation. Unlike traditional ARIMA-based models, TBATS was designed to handle time series with multiple seasonalities, long seasonal periods, and non-linear trends—making it highly applicable in healthcare demand forecasting scenarios. For example, TBATS could simultaneously account for daily, weekly, and yearly cycles in patient arrivals, which proved valuable in settings where hospital usage patterns reflected both short-term fluctuations and long-term periodicities.
[0006] Parallel to the evolution of statistical models, the rise of machine learning methods revolutionized the way researchers approached hospital queue prediction. Multilinear regression, one of the earliest and most widely applied methods, offered a framework for incorporating external covariates into forecasting models. While time series models like ARIMA and TBATS rely heavily on past values of the dependent variable, regression-based approaches allow integration of explanatory variables such as weather conditions, socio-economic indicators, public health announcements, or demographic data. In hospital contexts, multilinear regression has been applied to capture correlations between patient inflows and factors like temperature fluctuations, influenza outbreaks, or public holidays. This flexibility to include exogenous variables made regression-based models a critical complement to purely time series methods.
[0007] Beyond regression, machine learning introduced more sophisticated models such as random forests, support vector machines, and deep learning architectures, which were increasingly applied to patient flow prediction. These methods demonstrated strong predictive performance in many domains, but their "black box" nature often posed challenges in healthcare, where interpretability is as important as accuracy. Healthcare administrators and clinicians typically demand transparent models that allow them to understand the underlying drivers of patient arrivals. Therefore, hybrid approaches that combine statistical rigor with machine learning flexibility have emerged as promising avenues of research.
[0008] Globally, the problem of hospital overcrowding and unpredictable patient arrivals has been exacerbated by demographic and societal changes. Aging populations, increasing prevalence of chronic diseases, urbanization, and healthcare inequities all contribute to surging demand for medical services. Emergency departments, in particular, face overwhelming patient inflows, often leading to ambulance diversions, delayed treatments, and adverse patient outcomes. Several studies highlight that prolonged waiting times not only cause patient dissatisfaction but also increase risks of morbidity and mortality, especially in critical care scenarios. The COVID-19 pandemic further underscored the urgency of accurate demand forecasting, as hospitals around the world faced sudden surges in patient arrivals that exceeded their operational capacity. Forecasting models became essential tools for crisis management, enabling hospitals to prepare staffing schedules, allocate resources, and optimize triage protocols.
[0009] From a technological standpoint, advances in computational power and data availability have further facilitated the application of hybrid models for hospital queue prediction. The proliferation of electronic health records (EHRs), real-time monitoring systems, and health information exchanges has enabled researchers to access large volumes of patient arrival data. Cloud computing and distributed architectures have allowed for real-time processing of these data streams, while big data analytics tools provide the infrastructure for integrating heterogeneous sources of information. These developments have created fertile ground for sophisticated hybrid models that blend statistical and machine learning techniques for more accurate, dynamic, and context-sensitive forecasting of hospital queues.
[0010] At the same time, hospital management systems are increasingly adopting predictive analytics as a core operational tool. Decision support systems (DSS) now integrate forecasting modules to assist administrators in managing patient inflows and queues. These systems leverage predictive models not only for short-term scheduling but also for long-term capacity planning, staff allocation, and financial budgeting. The incorporation of hybrid models like TBATS and multilinear regression in such systems allows for both time-series based trend analysis and the inclusion of external influencing factors, thereby enhancing predictive robustness.
[0011] The academic literature is replete with examples of attempts to forecast hospital queues using a wide range of methodologies. Some studies employ purely statistical approaches, while others experiment with machine learning or hybrid models. However, a recurring theme across this body of work is the need to balance predictive accuracy with interpretability and computational feasibility. Purely statistical models like ARIMA are often interpretable but lack flexibility in handling multiple seasonalities. On the other hand, complex machine learning models may offer superior predictive performance but suffer from opacity and overfitting risks. The quest for hybrid approaches seeks to reconcile these trade-offs, leveraging the strengths of different methodologies while mitigating their weaknesses.
[0012] Another key consideration in the background of hospital queue forecasting is the diversity of application contexts. For instance, emergency departments exhibit highly volatile arrival patterns that require models with high responsiveness to sudden spikes. Outpatient clinics, by contrast, may display more regular and predictable patterns influenced by scheduled appointments and follow-up visits. Similarly, diagnostic units such as radiology or pathology labs have unique inflow characteristics shaped by referral patterns from other hospital departments. Forecasting models must therefore be adaptable to these diverse contexts while maintaining a consistent framework for data processing and prediction.
[0013] In addition, the social and economic implications of hospital queue forecasting cannot be overlooked. Long waiting times have been consistently linked to reduced patient satisfaction, which in turn affects hospital reputation and funding. In public health systems, overcrowded hospitals may lead to inequitable access to healthcare, disproportionately affecting vulnerable populations. From an economic standpoint, mismanagement of patient inflows can result in significant financial inefficiencies, including underutilized staff and equipment, increased overtime costs, and avoidable delays in service provision. Accurate forecasting of hospital queues thus carries profound implications for both patient welfare and healthcare economics.
[0014] Moreover, as hospitals increasingly operate in interconnected healthcare ecosystems, queue forecasting extends beyond individual institutions to broader networks of care. Regional health systems, for example, require coordinated forecasting models that can anticipate patient transfers, referral patterns, and resource sharing across multiple hospitals. This interconnectedness highlights the need for models that can capture both micro-level dynamics within a single hospital and macro-level interactions across healthcare networks.
[0015] In recent years, researchers have begun integrating hybrid time series and regression models into real-time hospital management platforms. These platforms not only forecast patient arrivals but also provide actionable insights for administrators. For instance, if a model predicts a surge in emergency department arrivals, the system can trigger pre-emptive measures such as mobilizing additional staff, preparing overflow facilities, or coordinating with nearby hospitals. Such real-time applications underscore the practical utility of advanced forecasting models in addressing the perennial challenges of hospital queue management.
[0016] Thus, in light of the above-stated discussion, there exists a need for a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting.
SUMMARY OF THE DISCLOSURE
[0017] The following is a summary description of illustrative embodiments of the invention. It is provided as a preface to assist those skilled in the art to more rapidly assimilate the detailed design discussion which ensues and is not intended in any way to limit the scope of the claims which are appended hereto in order to particularly point out the invention.
[0018] According to illustrative embodiments, the present disclosure focuses on a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting which overcomes the above-mentioned disadvantages or provide the users with a useful or commercial choice.
[0019] An objective of the present disclosure is to contribute towards efficient healthcare delivery systems by bridging the gap between statistical modeling and practical hospital management, ultimately ensuring optimized patient flow and improved clinical outcomes.
[0020] Another objective of the present disclosure is to develop a hybrid forecasting model that integrates TBATS (Trigonometric, Box-Cox, ARMA, Trend, and Seasonal components) with multilinear regression to effectively capture multi-seasonal trends in patient arrival patterns.
[0021] Another objective of the present disclosure is to enhance the accuracy of patient arrival forecasting by addressing limitations of traditional models like ARIMA and standalone regression, which often fail in handling complex hourly, daily, and weekly fluctuations.
[0022] Another objective of the present disclosure is to design a real-time queue prediction system that enables hospitals to anticipate queue length variations dynamically and plan resources accordingly.
[0023] Another objective of the present disclosure is to improve hospital resource management by providing actionable predictions that support proactive staffing, scheduling, and space allocation.
[0024] Another objective of the present disclosure is to reduce overcrowding and waiting times through precise forecasting of peak patient arrival periods, thereby improving patient satisfaction and service quality.
[0025] Another objective of the present disclosure is to facilitate adaptive decision-making by enabling administrators to monitor queue trends continuously and adjust operational strategies as needed.
[0026] Another objective of the present disclosure is to integrate predictive analytics into hospital workflows in a manner that supports automation and minimizes reliance on manual estimation methods.
[0027] Another objective of the present disclosure is to ensure scalability and flexibility of the proposed system so it can be applied to hospitals of varying sizes, patient demographics, and service capacities.
[0028] Yet another objective of the present disclosure is to validate the hybrid model’s performance against existing forecasting methods using real-world hospital datasets, ensuring robustness and reliability.
[0029] In light of the above, a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting comprises a data acquisition module configured to receive and store historical hospital arrival data. The system also includes a TBATS-based forecasting module configured to analyze the acquired data to detect and model multi-seasonal trends. The system also includes a multilinear regression module configured to integrate external influencing factors. The system also includes a hybrid prediction engine configured to generate queue length predictions and expected waiting time estimations for individual hospital departments. The system also includes an alert generation module configured to notify hospital administrators when predicted patient arrivals or queue lengths are forecasted to exceed department capacity thresholds.
[0030] In one embodiment, the data acquisition module is further configured to retrieve hospital arrival data from at least one of an electronic health record (EHR) system, hospital information system, or external data sources in real-time.
[0031] In one embodiment, the TBATS-based forecasting module is trained to capture multiple seasonalities including at least hourly, daily, and weekly variations in patient arrivals.
[0032] In one embodiment, the TBATS-based forecasting module applies a Box-Cox transformation to stabilize variance in patient arrival data prior to trend and seasonal analysis.
[0033] In one embodiment, the multilinear regression module incorporates external factors including at least holidays, weather conditions, epidemic alerts, and hospital workforce levels to refine patient queue predictions.
[0034] In one embodiment, the hybrid prediction engine combines outputs of the TBATS-based forecasting module and the multilinear regression module using a weighted ensemble approach to improve accuracy of queue length predictions.
[0035] In one embodiment, the hybrid prediction engine is configured to continuously update queue predictions based on incoming real-time patient arrival data.
[0036] In one embodiment, the hybrid prediction engine is configured to estimate expected patient waiting times by correlating predicted queue lengths with available workforce capacity.
[0037] In one embodiment, the alert generation module is configured to provide notifications via at least one of SMS, email, hospital dashboard, or mobile application when patient arrivals are forecasted to exceed department capacity thresholds.
[0038] In one embodiment, the alert generation module further prioritizes alerts based on department triage levels, assigning higher priority notifications to emergency and critical care units.
[0039] These and other advantages will be apparent from the present application of the embodiments described herein.
[0040] The preceding is a simplified summary to provide an understanding of some embodiments of the present invention. This summary is neither an extensive nor exhaustive overview of the present invention and its various embodiments. The summary presents selected concepts of the embodiments of the present invention in a simplified form as an introduction to the more detailed description presented below. As will be appreciated, other embodiments of the present invention are possible utilizing, alone or in combination, one or more of the features set forth above or described in detail below.
[0041] These elements, together with the other aspects of the present disclosure and various features are pointed out with particularity in the claims annexed hereto and form a part of the present disclosure. For a better understanding of the present disclosure, its operating advantages, and the specified object attained by its uses, reference should be made to the accompanying drawings and descriptive matter in which there are illustrated exemplary embodiments of the present disclosure.
BRIEF DESCRIPTION OF THE DRAWINGS
[0042] To describe the technical solutions in the embodiments of the present disclosure or in the prior art more clearly, the following briefly describes the accompanying drawings required for describing the embodiments or the prior art. Apparently, the accompanying drawings in the following description merely show some embodiments of the present disclosure, and a person of ordinary skill in the art can derive other implementations from these accompanying drawings without creative efforts. All of the embodiments or the implementations shall fall within the protection scope of the present disclosure.
[0043] The advantages and features of the present disclosure will become better understood with reference to the following detailed description taken in conjunction with the accompanying drawing, in which:
[0044] FIG. 1 illustrates a flowchart outlining sequential step involved in a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting, in accordance with an exemplary embodiment of the present disclosure;
[0045] FIG. 2 illustrates a flowchart showing working of a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting, in accordance with an exemplary embodiment of the present disclosure.
[0046] Like reference, numerals refer to like parts throughout the description of several views of the drawing;
[0047] The hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting, which like reference letters indicate corresponding parts in the various figures. It should be noted that the accompanying figure is intended to present illustrations of exemplary embodiments of the present disclosure. This figure is not intended to limit the scope of the present disclosure. It should also be noted that the accompanying figure is not necessarily drawn to scale.
DETAILED DESCRIPTION OF THE DISCLOSURE
[0048] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present disclosure.
[0049] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It may be apparent to one skilled in the art that embodiments of the present disclosure may be practiced without some of these specific details.
[0050] Various terms as used herein are shown below. To the extent a term is used, it should be given the broadest definition persons in the pertinent art have given that term as reflected in printed publications and issued patents at the time of filing.
[0051] The terms “a” and “an” herein do not denote a limitation of quantity, but rather denote the presence of at least one of the referenced items.
[0052] The terms “having”, “comprising”, “including”, and variations thereof signify the presence of a component.
[0053] Referring now to FIG. 1 to FIG. 2 to describe various exemplary embodiments of the present disclosure. FIG. 1 illustrates a flowchart outlining sequential step involved in a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting, in accordance with an exemplary embodiment of the present disclosure.
[0054] A hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting 100 comprises a data acquisition module 102 configured to receive and store historical hospital arrival data. The data acquisition module 102 is further configured to retrieve hospital arrival data from at least one of an electronic health record (EHR) system, hospital information system, or external data sources in real-time.
[0055] The system also includes a TBATS-based forecasting module 104 configured to analyze the acquired data to detect and model multi-seasonal trends. The TBATS-based forecasting module 104 is trained to capture multiple seasonalities including at least hourly, daily, and weekly variations in patient arrivals. The TBATS-based forecasting module 104 applies a Box-Cox transformation to stabilize variance in patient arrival data prior to trend and seasonal analysis.
[0056] The system also includes a multilinear regression module 106 configured to integrate external influencing factors. The multilinear regression module 106 incorporates external factors including at least holidays, weather conditions, epidemic alerts, and hospital workforce levels to refine patient queue predictions.
[0057] The system also includes a hybrid prediction engine 108 configured to generate queue length predictions and expected waiting time estimations for individual hospital departments. The hybrid prediction engine 108 combines outputs of the TBATS-based forecasting module and the multilinear regression module using a weighted ensemble approach to improve accuracy of queue length predictions. The hybrid prediction engine 108 is configured to continuously update queue predictions based on incoming real-time patient arrival data. The hybrid prediction engine 108 is configured to estimate expected patient waiting times by correlating predicted queue lengths with available workforce capacity.
[0058] The system also includes an alert generation module 110 configured to notify hospital administrators when predicted patient arrivals or queue lengths are forecasted to exceed department capacity thresholds. The alert generation module 110 is configured to provide notifications via at least one of SMS, email, hospital dashboard, or mobile application when patient arrivals are forecasted to exceed department capacity thresholds. The alert generation module 110 further prioritizes alerts based on department triage levels, assigning higher priority notifications to emergency and critical care units.
[0059] FIG. 1 illustrates a flowchart outlining sequential step involved in a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting.
[0060] At 102 the process begins with the data acquisition module, which is responsible for receiving and storing historical hospital arrival data. This data may include detailed records of patient arrivals over time, categorized by parameters such as admission date, arrival time, department allocation, and triage level. By compiling such historical information, the system develops a robust foundation upon which forecasting models can be applied. These data inputs are sourced from hospital databases, electronic health records (EHRs), or administrative Excel sheets, ensuring a comprehensive view of patient arrival patterns over different time horizons.
[0061] At 104, once data acquisition is complete, the system activates the TBATS-based forecasting module, which is configured to analyze the historical data and detect recurring seasonal trends in patient arrivals. TBATS, which incorporates Trigonometric functions, Box-Cox transformations, ARMA errors, and trend-seasonal decomposition, is particularly suited for environments like hospitals where arrival patterns exhibit multi-seasonality. For example, patient arrivals may follow daily cycles with morning and evening peaks, weekly cycles with variations on weekdays and weekends, and even seasonal patterns such as increased arrivals during flu season. The TBATS module is trained on this diverse data to identify and model such variations, thereby enabling accurate forecasting of baseline patient inflows that are influenced by intrinsic temporal rhythms.
[0062] At 106, parallel to the TBATS analysis, the system incorporates the multilinear regression module, which accounts for external influencing factors not captured solely by historical arrival patterns. Multilinear regression integrates outbound determinants such as public holidays, extreme weather conditions, or variations in workforce availability, each of which significantly impacts hospital inflows and queue lengths. For instance, a sudden drop in workforce availability during staff strikes or increased patient arrivals during regional festivals can disrupt standard patterns. By mathematically modeling these variables alongside the TBATS outputs, the regression module provides contextual adjustments that enhance prediction accuracy.
[0063] At 108, the outputs from both the TBATS-based forecasting module and the multilinear regression module are integrated within the hybrid prediction engine. This engine synthesizes multi-seasonal time-series predictions with regression-based contextual influences to generate highly refined forecasts. The hybrid prediction engine is designed to produce two crucial outputs: predicted queue lengths in various hospital departments and estimations of expected patient waiting times. These outputs are highly actionable, as they directly translate predictive insights into operational measures, offering hospital administrators visibility into future demand and enabling them to prepare accordingly.
[0064] At 110, to ensure proactive management, the system also incorporates an alert generation module. This module constantly monitors predicted patient arrivals and queue lengths against established hospital capacity thresholds. When the forecast indicates that incoming patient volumes or waiting times are likely to exceed departmental capacity limits, the alert generation module issues warnings to hospital administrators. These notifications act as early signals, allowing administrators to respond before congestion escalates into critical delays.
[0065] Once alerts are generated, the computed predictions are compared with the hospital’s existing capacity to evaluate potential mismatches. If the load is projected to exceed capacity, the system supports administrative decision-making by providing quantitative insights into anticipated patient pressure. Administrators can then make informed choices regarding resource allocation, such as reassigning medical staff, adjusting shift timings, or deploying additional equipment to departments under stress. This dynamic response mechanism ensures that patient flow across hospital departments is streamlined, ultimately reducing waiting times and improving care quality.
[0066] FIG. 2 illustrates a flowchart showing working of a hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting.
[0067] The process begins with the collection of historical hospital arrival data, which serves as the foundation for the predictive modeling. These data sources may originate from hospital Excel records or electronic health records (EHRs), ensuring that relevant past information is systematically captured. The acquired data is analyzed with respect to critical attributes such as patient arrival date, time of entry, departmental allocation, and triage level, as these parameters directly influence hospital load and patient flow.
[0068] Once the foundational data is analyzed, the system proceeds into two parallel processing streams. In the first stream, the TBATS model is trained on the extracted data. TBATS, which stands for Trigonometric, Box-Cox, ARMA, Trend, and Seasonal model, is particularly suited for identifying and learning multi-seasonal patterns that characterize hospital arrivals, such as hourly peaks, daily variations, and weekly cycles. This ensures that repetitive and seasonal fluctuations in patient inflow are accurately captured. In the second stream, the system incorporates multilinear regression (MLR) by accounting for external influencing factors that may not be inherent in the hospital’s internal records. These outbound factors include holidays, weather conditions, and workforce availability, all of which significantly impact patient arrival rates and queue build-up.
[0069] After both TBATS modeling and MLR factor incorporation, the system merges insights to establish refined predictions. The TBATS output contributes an understanding of recurring multi-seasonal trends, while the MLR module adds contextual real-world factors that influence patient load dynamics. The integration of these two approaches allows the system to predict queue lengths across hospital departments and estimate expected waiting times with high accuracy. This hybrid prediction output not only anticipates routine inflow but also adjusts to irregular circumstances that might otherwise disrupt patient flow management.
[0070] The subsequent stage of the process involves proactive monitoring. When predicted queue lengths or waiting times indicate high patient load conditions, the system generates warnings for administrators, highlighting the possibility of demand exceeding the hospital’s operational capacity. This leads to a computed evaluation of load versus capacity, comparing anticipated patient inflow against available resources such as medical staff, equipment, and beds. With this comparative analysis, hospital decision-makers are empowered to intervene effectively.
[0071] Based on the forecasted outcomes, the system supports informed decision-making processes. Administrators can use these insights to optimize resource allocation, ensuring that medical staff and infrastructure are deployed in accordance with predicted demand. By doing so, bottlenecks in hospital operations are minimized, and patients experience reduced waiting times. Ultimately, the system facilitates improved patient flow throughout departments by dynamically aligning available resources with anticipated needs.
[0072] The entire process is cyclical and adaptive, allowing the system to continuously refine its predictions with incoming data. By combining statistical time-series modeling with regression-based context analysis, this hybrid framework not only forecasts patient arrivals but also provides actionable intelligence to improve hospital management. The final outcome is an optimized healthcare environment where resources are balanced with patient demand, leading to enhanced operational efficiency and improved patient care delivery.
[0073] While the invention has been described in connection with what is presently considered to be the most practical and various embodiments, it will be understood that the invention is not to be limited to the disclosed embodiments, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the scope of the appended claims.
[0074] A person of ordinary skill in the art may be aware that, in combination with the examples described in the embodiments disclosed in this specification, units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof.
[0075] The foregoing descriptions of specific embodiments of the present disclosure have been presented for purposes of illustration and description. They are not intended to be exhaustive or to limit the present disclosure to the precise forms disclosed, and many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described to best explain the principles of the present disclosure and its practical application, and to thereby enable others skilled in the art to best utilize the present disclosure and various embodiments with various modifications as are suited to the particular use contemplated. It is understood that various omissions and substitutions of equivalents are contemplated as circumstances may suggest or render expedient, but such omissions and substitutions are intended to cover the application or implementation without departing from the scope of the present disclosure.
[0076] Disjunctive language such as the phrase “at least one of X, Y, Z,” unless specifically stated otherwise, is otherwise understood with the context as used in general to present that an item, term, etc., may be either X, Y, or Z, or any combination thereof (e.g., X, Y, and/or Z). Thus, such disjunctive language is not generally intended to, and should not, imply that certain embodiments require at least one of X, at least one of Y, or at least one of Z to each be present.
[0077] In a case that no conflict occurs, the embodiments in the present disclosure and the features in the embodiments may be mutually combined. The foregoing descriptions are merely specific implementations of the present disclosure, but are not intended to limit the protection scope of the present disclosure. Any variation or replacement readily figured out by a person skilled in the art within the technical scope disclosed in the present disclosure shall fall within the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.
, Claims:I/We Claim:
1. A hybrid hospital queue prediction system using TBATS and multilinear regression for dynamic patient arrival forecasting (100) comprising:
a data acquisition module (102) configured to receive and store historical hospital arrival data;
a TBATS-based forecasting module (104) configured to analyze the acquired data to detect and model multi-seasonal trends;
a multilinear regression module (106) configured to integrate external influencing factors;
a hybrid prediction engine (108) configured to generate queue length predictions and expected waiting time estimations for individual hospital departments;
an alert generation module (110) configured to notify hospital administrators when predicted patient arrivals or queue lengths are forecasted to exceed department capacity thresholds.
2. The system (100) as claimed in claim 1, wherein the data acquisition module (102) is further configured to retrieve hospital arrival data from at least one of an electronic health record (EHR) system, hospital information system, or external data sources in real-time.
3. The system (100) as claimed in claim 1, wherein the TBATS-based forecasting module (104) is trained to capture multiple seasonalities including at least hourly, daily, and weekly variations in patient arrivals.
4. The system (100) as claimed in claim 1, wherein the TBATS-based forecasting module (104) applies a Box-Cox transformation to stabilize variance in patient arrival data prior to trend and seasonal analysis.
5. The system (100) as claimed in claim 1, wherein the multilinear regression module (106) incorporates external factors including at least holidays, weather conditions, epidemic alerts, and hospital workforce levels to refine patient queue predictions.
6. The system (100) as claimed in claim 1, wherein the hybrid prediction engine (108) combines outputs of the TBATS-based forecasting module and the multilinear regression module using a weighted ensemble approach to improve accuracy of queue length predictions.
7. The system (100) as claimed in claim 1, wherein the hybrid prediction engine (108) is configured to continuously update queue predictions based on incoming real-time patient arrival data.
8. The system (100) as claimed in claim 1, wherein the hybrid prediction engine (108) is configured to estimate expected patient waiting times by correlating predicted queue lengths with available workforce capacity.
9. The system (100) as claimed in claim 1, wherein the alert generation module (110) is configured to provide notifications via at least one of SMS, email, hospital dashboard, or mobile application when patient arrivals are forecasted to exceed department capacity thresholds.
10. The system (100) as claimed in claim 1, wherein the alert generation module (110) further prioritizes alerts based on department triage levels, assigning higher priority notifications to emergency and critical care units.