Context-aware scheduling and allocation system

Context-aware scheduling and allocation systemCASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Phase 1: Information flow, workflow and patient pathways involved in the scheduling process Data collection through individual semi-structured interviews and observations. It is planned about 20 interviews with patients and health personnel, and observations of the workflow at the hospital. Common work patterns will be modelled as input to Phase 2, 3 and 4. The qualitative approach focus on the interplay between technical and social factors that produces particular outcomes. Phase 2: Adaptive workflow modelling Comprehensive and systematic data collection, both manually and automatically, from the health information systems involved in the scheduling process. These data will be required in Phase 3 to generate adaptive workflows, and in Phase 4 as input for CMS. Phase 3: Simulation technique for patient appointment scheduling Phase 4: Optimization algorithm applied to adaptive workflows Identify models for pattern recognition and machine learning that will allow the context engine (CMS) to identify work patterns. Phase 5: System evaluation and overall recommendations Compare the workflow, resource usage and service capacity before and after use of the developed system prototype. Assign costs to each major operational change as described in 5.1 in order to compare resulting operational expenditures before and after introducing the system prototype. The tasks in Phase 1 and Phase 2 that require interviews and observations of patients and clinical staff at UNN had to be postponed due to: 1) sabbatical leave of the contact person at the surgical department in the first year; 2) From March 2020 till spring 2022 the access to patients and clinical staff was not possible due to the Covid-19 pandemic, and thereby it has been impossible to plan for interviews and observations at the hospital. The researcher has been on sick leave and maternity leave since spring 2022. Due to the reallocation of the resource due to COVID, it is unknown when it will be possible to re-start the above mentioned tasks in Phase 1 and 2, we propose to postpone the project until it is possible to complete these tasks.

Context-aware scheduling and allocation systemCASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Phase 1: Information flow, workflow and patient pathways involved in the scheduling process Data collection through individual semi-structured interviews and observations. It is planned about 20 interviews with patients and health personnel, and observations of the workflow at the hospital. Common work patterns will be modelled as input to Phase 2, 3 and 4. The qualitative approach focus on the interplay between technical and social factors that produces particular outcomes. Phase 2: Adaptive workflow modelling Comprehensive and systematic data collection, both manually and automatically, from the health information systems involved in the scheduling process. These data will be required in Phase 3 to generate adaptive workflows, and in Phase 4 as input for CMS. Phase 3: Simulation technique for patient appointment scheduling Phase 4: Optimization algorithm applied to adaptive workflows Identify models for pattern recognition and machine learning that will allow the context engine (CMS) to identify work patterns. Phase 5: System evaluation and overall recommendations Compare the workflow, resource usage and service capacity before and after use of the developed system prototype. Assign costs to each major operational change as described in 5.1 in order to compare resulting operational expenditures before and after introducing the system prototype. The tasks in Phase 1 and Phase 2 that require interviews and observations of patients and clinical staff at UNN had to be postponed due to: 1) sabbatical leave of the contact person at the surgical department in the first year; 2) From March 2020 till spring 2022 the access to patients and clinical staff was not possible due to the Covid-19 pandemic, and thereby it has been impossible to plan for interviews and observations at the hospital. The researcher has been on sick leave and maternity leave since spring 2022. Due to the reallocation of the resource due to COVID, it is unknown when it will be possible to re-start the above mentioned tasks in Phase 1 and 2, we propose to postpone the project until it is possible to complete these tasks.

Context-aware scheduling and allocation system (CASAS)CASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Phase 1: Information flow, workflow and patient pathways involved in the scheduling process Data collection through individual semi-structured interviews and observations. It is planned about 20 interviews with patients and health personnel, and observations of the workflow at the hospital. Common work patterns will be modelled as input to Phase 2, 3 and 4. The qualitative approach focus on the interplay between technical and social factors that produces particular outcomes. Phase 2: Adaptive workflow modelling Comprehensive and systematic data collection, both manually and automatically, from the health information systems involved in the scheduling process. These data will be required in Phase 3 to generate adaptive workflows, and in Phase 4 as input for CMS. Phase 3: Simulation technique for patient appointment scheduling Phase 4: Optimization algorithm applied to adaptive workflows Identify models for pattern recognition and machine learning that will allow the context engine (CMS) to identify work patterns. Phase 5: System evaluation and overall recommendations Compare the workflow, resource usage and service capacity before and after use of the developed system prototype. Assign costs to each major operational change as described in 5.1 in order to compare resulting operational expenditures before and after introducing the system prototype. The tasks in Phase 1 and Phase 2 that require interviews and observations of patients and clinical staff at UNN had to be postponed due to: 1) sabbatical leave of the contact person at the surgical department in the first year; 2) In March 2020 the access to patients and clinical staff was not possible due to the ongoing Covid-19 pandemic. This situation did not change in 2021 and thereby it is impossible to plan for interviews and observations at the hospital. Since it is unknown when it will be possible to re-start the above mentioned tasks in Phase 1 and 2, we propose to postpone the project until it is possible to complete these tasks.

Context-aware scheduling and allocation system (CASAS)CASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Phase 1: Information flow, workflow and patient pathways involved in the scheduling process Data collection through individual semi-structured interviews and observations. It is planned about 20 interviews with patients and health personnel, and observations of the workflow at the hospital. Common work patterns will be modelled as input to Phase 2, 3 and 4. The qualitative approach focus on the interplay between technical and social factors that produces particular outcomes. Phase 2: Adaptive workflow modelling Comprehensive and systematic data collection, both manually and automatically, from the health information systems involved in the scheduling process. These data will be required in Phase 3 to generate adaptive workflows, and in Phase 4 as input for CMS. Phase 3: Simulation technique for patient appointment scheduling Phase 4: Optimization algorithm applied to adaptive workflows Identify models for pattern recognition and machine learning that will allow the context engine (CMS) to identify work patterns. Phase 5: System evaluation and overall recommendations Compare the workflow, resource usage and service capacity before and after use of the developed system prototype. Assign costs to each major operational change as described in 5.1 in order to compare resulting operational expenditures before and after introducing the system prototype. The tasks in Phase 1 and Phase 2 that require interviews and observations of patients and clinical staff at UNN had to be postponed due to: 1) sabbatical leave of the contact person at the surgical department in the first year; 2) In March 2020 the access to patients and clinical staff was not possible due to the ongoing Covid-19 pandemic. This situation is not likely to change in 2021 and thereby it is impossible to plan for interviews and observations at the hospital. Since it is unknown when it will be possible to re-start the above mentioned tasks in Phase 1 and 2, we propose to change these tasks to a theoretical approach in order to avoid delays in the ending date of the project.

Context-aware scheduling and allocation system (CASAS)CASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Surgical departments are simultaneously the major source of investment, and the greatest source of revenue for most hospitals. However, it is known that between 10 and 40 % of elective surgeries are cancelled. In western countries, up to 20 % of elective surgeries are cancelled on the day of surgery. Surgery cancellations are undesirable in hospital settings as they increase costs, reduce productivity and efficiency, increase waiting lists, and directly affect the patient. Considerable resources are invested in maintaining operating theatres, and having surgeons and theatre staff available on an agreed schedule. The majority of the hospitals use a so-called block-booking system when planning surgeries. In this system, a medical specialty is assigned to blocks denoting a specific amount of time, e.g., a day, in one Operating Room (OR). These blocks can be combined into cyclical Master Surgery Schedules (MSS), where every block is repeated after a fixed cycle. At the strategic level of block-booking system, the number of blocks assigned to the specialties and emergencies during a MSS cycle is determined. At the tactical level, OR-days are allocated to specialties in an MSS, such that the strategic allocation is met. Uncertainties might come from different sources, such as processing times, demand/patient arrivals, no-show ups, personnel availability, etc. Clearly, MSS affects the patient flow to downstream inpatient care units. The development of an MSS module must address three main challenges: 1) Enlarging the scope of the MSS: MSS approaches embedded in commercial software consider only the impact of the MSS on operating theatre and operating staff; the goal here is to enlarge the scope to down-stream resources, such as the intensive care unit ICU and the general wards required by the patients. The solution module should be flexible enough to cope with different features that appear in different hospitals that interfere with the planning activities. 2) Planning with uncertainty: Surgical management processes are subject to high variability resulting in significant deviations between intended and actual performance of surgical plans. For instance, when surgeries take longer than predicted or emergency patients arrive, it often results in overtime and possible cancellation of surgeries. When planning at an aggregate level, uncertainties are usually neglected. The challenge is to anticipate the uncertainties and incorporate them during the MSS decision-making. 3) Solution approaches: The problem cannot be totally described in mathematical programming terms. The volatility of information (see previous point) makes it difficult to incorporate all uncertainty in a single solid deterministic model. To tackle such challenges, a MMS module has to enable a fast and automated, fully context dependent, scheduling. In such scenario, context-aware systems present themselves as a promising approach. This first year was used to startup the project, initiating with the conduction of a systematic review on the challenges arising from e-health interventions with focus on the workflow related factors, leading to the publication of a level 2 scientific article. Foreseeing future work, different optimization and metaheuristic algorithms were studied, and the most suitable for the problem in question identified. This year was also marked by early stage development of a simulation technique for patient appointment scheduling and resource allocation.

Context-aware scheduling and allocation system (CASAS)CASAS studies the use of context-based data in the scheduling of elective surgeries. Through a comprehensive characterization of the providers operations, the hypothesis is that a better adjusted patient appointment schedule will contribute to an efficient use of existing resources, and a reduction in the number of elective surgery cancellations.Surgical departments are simultaneously the major source of investment, and the greatest source of revenue for most hospitals. However, it is known that between 10 and 40 % of elective surgeries are cancelled. In western countries, up to 20 % of elective surgeries are cancelled on the day of surgery. Surgery cancellations are undesirable in hospital settings as they increase costs, reduce productivity and efficiency, increase waiting lists, and directly affect the patient. Considerable resources are invested in maintaining operating theatres, and having surgeons and theatre staff available on an agreed schedule. The majority of the hospitals use a so-called block-booking system when planning surgeries. In this system, a medical specialty is assigned to blocks denoting a specific amount of time, e.g., a day, in one Operating Room (OR). These blocks can be combined into cyclical Master Surgery Schedules (MSS), where every block is repeated after a fixed cycle. At the strategic level of block-booking system, the number of blocks assigned to the specialties and emergencies during a MSS cycle is determined. At the tactical level, OR-days are allocated to specialties in an MSS, such that the strategic allocation is met. Uncertainties might come from different sources, such as processing times, demand/patient arrivals, no-show ups, personnel availability, etc. Clearly, MSS affects the patient flow to downstream inpatient care units. The development of an MSS module must address three main challenges: 1) Enlarging the scope of the MSS: MSS approaches embedded in commercial software consider only the impact of the MSS on operating theatre and operating staff; the goal here is to enlarge the scope to down-stream resources, such as the intensive care unit ICU and the general wards required by the patients. The solution module should be flexible enough to cope with different features that appear in different hospitals that interfere with the planning activities. 2) Planning with uncertainty: Surgical management processes are subject to high variability resulting in significant deviations between intended and actual performance of surgical plans. For instance, when surgeries take longer than predicted or emergency patients arrive, it often results in overtime and possible cancellation of surgeries. When planning at an aggregate level, uncertainties are usually neglected. The challenge is to anticipate the uncertainties and incorporate them during the MSS decision-making. 3) Solution approaches: The problem cannot be totally described in mathematical programming terms. The volatility of information (see previous point) makes it difficult to incorporate all uncertainty in a single solid deterministic model. To tackle such challenges, a MMS module has to enable a fast and automated, fully context dependent, scheduling. In such scenario, context-aware systems present themselves as a promising approach. This first year was used to startup the project, initiating with the conduction of a systematic review on the challenges arising from e-health interventions with focus on the workflow related factors, leading to the publication of a level 2 scientific article. Foreseeing future work, different optimization and metaheuristic algorithms were studied, and the most suitable for the problem in question identified. This year was also marked by early stage development of a simulation technique for patient appointment scheduling and resource allocation.