• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of arranging holdings with the following steps: keeping a history of performed holdings; keeping an agenda of availability; determining the practitioners to perform a hold at a particular time within the region taking into account the population of the region to be served, determining a number of possible practitioners for each hold and assigning the hold times to be performed, taking into account taking into account the geographical spread, the availability and the waiting times already performed.
    公开号:BE1019491A5
    申请号:E2011/0337
    申请日:2011-05-31
    公开日:2012-07-03
    发明作者:Freek Rombauts
    申请人:Konink Apothekersvereniging Van Antwerpen；
    IPC主号:
    专利说明:

    A method of controlling queues within a group of practitioners of the same profession for a predetermined region and a computer program in a computer readable format, comprising software code portions for performing this method
    This invention relates to a method for controlling queuing within a group of practitioners of the same profession for a predetermined region.
    The invention also relates to a computer program in a computer readable format, comprising software code portions for performing the method according to the invention.
    It is known that for some professions, such as pharmacists, general practitioners, etc., it is desirable or legally required that a so-called watch service is organized. The purpose of such a watch service is to ensure continuity in the availability of these professions. Examples of this can be found, for example, in CA2593371A1 and FREY et al .: "Optimizing staff rosters for emergency shifts for doctors", IEEE CONGRESS ON EVOLUTIONARY COMPUTATION, 2009, CEC Ό9, IEEE, PISCATAWAY, NJ, USA, May 18, 2009, pages 2540 -2546, XP031466534, ISBN: 978-1-4244-2958-5.
    However, organizing such a watch service is not easy. After all, the preferences of, for example, public holidays, leave, fairs, markets, roadworks in the vicinity of the practitioner, etc., of the various practitioners to perform their watch must be taken into account. It should also be taken into account that a sufficient number of practitioners are available in a certain region, depending on the population of the region and preferably also on the maximum travel time for the users of this watch service based on the population density.
    It is therefore an object of the present invention to provide a method that makes it possible to control waiting sessions within a group of practitioners.
    This object is achieved in accordance with the content of the first claim.
    To this end, the method comprises the following steps: - keeping a history of performed waiting times per practitioner; - keeping an agenda of availability for the waiting sessions per practitioner; - determining the practitioners who must perform a wait at a certain time within the region, taking into account the population of the region to be served, - determining a number of possible practitioners for each wait to be performed and assigning the wait Waitings to be performed on at least one of them, taking into account the geographic spread, the availability and the already performed waiting times.
    By effecting such a spread of the practitioners, it is achieved that a better geographical spread of the practitioners is achieved. This benefits the residents of the area as every resident is a guard in a more dispersed way. This is also advantageous for the practitioners themselves, since it is avoided that at less busy times (at night and / or during the weekend) practitioners are too close to each other, which can result in an unnecessary supply to the residents. Partly as a result, the waiting frequency (the frequency that a practitioner is on duty) decreases and the permanence of the practitioner is better utilized. Because a centralized procedure for distributing the waiting can be provided for, the administrative burden for practitioners is also reduced.
    In the context of the current patent application, wait is defined as the time span that a practitioner of a group of practitioners is on hold within a predetermined region. A waiting period is preferably determined by which part of the day (for example day or night) and / or which type (for example week or weekend). Related waiting parts, for example waiting parts of the same type, where waiting parts are day parts on specific days, can be combined. For example, a wait can include a week watch, which groups, for example, the day watch Monday, Tuesday, Wednesday, and Thursday, a weekend watch, which groups, for example, the watch, day watch Friday, Saturday, and Sunday, and seven night watches, which means the day part night of the different days .
    The profession relates to any profession related to services, such as for example the healthcare sector, but is preferably a profession where arranging watchings is desirable, necessary or legally required, such as, in particular, a medical profession such as, for example, a pharmacist, (general) doctor, dentist , surgeon, veterinarian, nurses, ambulance services, waiting arrangements within a hospital or within a group of hospitals, clinical labs, etc.
    The predetermined region within which the method controls waiting is also referred to as waiting region within the current patent application. For example, the group of practitioners of the profession within the predetermined region is arranged such that the group of practitioners of the profession collects practitioners within a certain administrative region. The size of the administrative region can be determined freely here. For example, it is possible to use a national, provincial, municipal, urban, etc. region, depending on, for example, the profession.
    In the context of the present patent application, practitioner of the profession is defined as one or more persons who are active within an administrative entity, such as, for example, a company, practice, etc., and who have the possibility of practicing the profession. For example, in the context of the current patent application, a practice where several practitioners, such as, for example, a few doctors and / or pharmacists are employed, is also referred to as a practitioner. In Belgium, a pharmacy practice is, for example, characterized by a unique APB (General Pharmaceutical Association) number.
    For example, a waiting area may represent a single day period for the group Monday through Thursday, from 9 a.m. to 10 p.m. (referred to as a day watch in the context of the current patent application), a watch for a single day period from Friday to Sunday and from Sunday to Friday from 9 a.m. to 10 p.m. (referred to in the context of the current patent application as weekend watch) or a watch for the single night period of the group Monday to Sunday from 10 p.m. until 9 a.m. (called night watch in the context of the current patent application). The determination of waiting periods can be further refined by taking into account public holidays. For example, a weekend watch contains official holidays if they fall on Monday or Thursday. In addition, a weekend watch preferably includes Thursday if the public holiday falls on a Friday. If a public holiday falls on a Thursday, the weekend watch preferably starts as always on a Friday. A waiting week collects all waiting times per week, and largely corresponds to a calendar week. A waiting year starts on the first Monday of the year and ends on the last day of the last week's watch that starts in that year. A waiting year therefore has 52 or 53 waiting weeks. With regard to standby duty of a public holiday practice, the total number of public holidays per year is preferably taken into account, even if they fall on Sundays. For example, if a practice has a duty on a public holiday, it will be lost for the next 9 years. A practice is therefore at most every 10 years of waiting for a holiday.
    According to the present invention, the method for controlling queuing is performed at least in part by at least one computer. Namely, in such an embodiment, a considerable acceleration can be realized in determining a good control of waiting times in the waiting region. The completion of the waiting year or part of it can be performed on the basis of one or more algorithms on the at least one computer.
    According to further preferred embodiments of the present invention, determining a number of possible practitioners for each wait to be performed, preferably taking into account the availability and the already performed waitings, is performed in a first step. For example, the practitioners selected in the first step for a particular wait are kept in a first list, preferably with a weight assigned to the different practitioners of the first list, taking into account the availability and the already performed waiting sessions. The assignment of the waiting to be performed to at least one of them, being the possible waiting practitioners determined in the first step, taking into account the availability and the already performed waiting, taking into account the geographic spread, is carried out in a subsequent second individual step. Preferably, the second step takes into account the weight assigned to the practitioners selected in the first step. For example, the practitioners finally selected are kept in a second list. In this way it is made possible for the first step to be performed by a first instance that typically has easy access to often personal information without having to pass on too much personal information to a second instance which can then perform the second step based on the results of the first step. For example, a professional association of practitioners, such as an association grouping pharmacists within a certain region, can carry out the first step and the second step can be carried out by a company that is more specialized in processing geographic information, without necessarily involving personal data from the practitioners need to be passed on to the company.
    Assigning the wait to one of the practitioners selected in the second step, taking into account their geographical spread, takes into account, for example, one or more of the following: - the geographical location of the practitioner - the population density of the waiting region - the coverage figure for the waiting area to be executed. The coverage figure can, for example, be between a minimum coverage figure for the waiting part to be performed (= minimum number of residents per practitioner), possibly depending on the individual capacity of practitioners and a maximum coverage figure for the waiting part to be performed (= maximum number of residents per practitioner) and is furthermore preferably also dependent on the individual capacity of practitioners and possibly also an optimal coverage figure for the waiting part to be carried out (= optimum number of inhabitants per practitioner), possibly dependent on the individual capacity of practitioners - minimum distance between two practitioners for guard during the waiting part, preferably this distance also depends on the coverage number - maximum distance between two practitioners for guard during the waiting part, preferably this distance also depends on the coverage number - minimum number of guard practitioners within the waiting region for the waiting part - the weight assigned by the first step to the different practitioners selected in the first step - a number of core areas, areas where a practitioner must be present regardless of the other parameters such as areas that are difficult for residents to leave, hospitals, psychological boundaries, etc.
    - practitioners who must be selected The second step is preferably done with the help of one or more algorithms. A first algorithm makes a selection from a series of candidates, taking into account the weight of the candidate practitioner and will remove the practitioners who are geographically too close to already selected practitioners. Then all practitioners are removed that are in a core area in which at least 1 practitioner has already been selected. In a next step, one practitioner is selected per core area without already selected practitioners. If this core area is adjacent to 1 or more core areas in which a practitioner has already been selected, the practitioner is selected with optimum distance from the selected practitioners, otherwise the practitioner with the largest weight is selected. Subsequently, all other practitioners are removed in the core area. Again, the practitioners are removed that are too close to the selected practitioners and the practitioners are selected outside the core areas. This is done, for example, by assigning the population within the area to the nearest practitioner on the basis of the “shortest path algorithm” and selecting the practitioner who is optimally located relative to the selected practitioner with the most population who does not border on the area from another selected practitioner. Finally, those selected practitioners are removed who, when removed, do not cause the maximum population of another selected practitioner to be exceeded. A second algorithm splits the candidates into 2 series that are equally well-spaced. Use is made of, for example, graph coloring. It has been found that such a method can be carried out in a relatively short time and yet gives a good spread.
    However, the second step can also be performed using a Voronoi algorithm using one or preferably several or more preferably all of the above-mentioned parameters.
    For example, it is possible to create a Voronoi distribution based on all the practitioners on the first list for the waiting area for the specific waiting area. After this distribution has been established, the number of inhabitants within each waiting area determined by the Voronoi distribution within the waiting region is calculated on the basis of population density. Subsequently, a selection value is determined for each sub-area by multiplying the number of inhabitants within the sub-area by a selection coefficient. Next, it is checked whether this selection value is within predetermined limit values and whether these practitioners also fulfill the other parameters, such as for example minimum distance between two practitioners for guard during the waiting part, maximum distance between two practitioners for guard during the waiting part, minimum number of guard practitioners within the waiting part waiting area for the waiting area. If this is not the case, some practitioners from the first list are omitted, preferably those with the lowest selection value. After this, a new Voronoi distribution is made on the basis of the shortened list and it is checked again whether the associated selection values are within the predetermined limit values and whether the other parameters are met. If this is not the case again, one or more practitioners can be added again after which others are removed from the list, one or more practitioners can be further removed from the list, etc. This process is then repeated until a second list is obtained that meets the various selection criteria. Although it would also be possible, for example, to create a Voronoi distribution for all different possible compositions of practitioners from the first list, the advantage of the strategy outlined above is that, although not necessarily the most ideal distribution will always be found, a good distribution is found within a shorter time.
    According to further preferred embodiments of the present invention, the second step is performed by means of a GIS or "Geography Information System" or in the Dutch "Geographic Information System". Such a system has the necessary tools to be able to smoothly carry out the second step on the basis of various parameters that can be entered in advance into the system and the information from the first step.
    According to further preferred embodiments of the present invention, the first and the second step are performed by a first and a second computer system, respectively. Such an implementation has the advantage that the first computer system, often operated by a first institution having access to the history of performed watchings, the availability agenda of the various practitioners, can operate independently of the second computer system, thus reducing the risk of sensitive information from the first instance to the second instance. In addition, a body such as the first one, often a professional association, is not obliged to develop a full GIS for relatively limited use, but it is possible that the GIS can be entrusted to the second one, which can therefore better focus on maintenance of the GIS, which is often very intensive, and can also use the GIS for other purposes.
    According to further preferred embodiments of the present invention, when assigning the waiting times to be performed to at least one of them, taking into account the geographic spread, a pre-set minimum and maximum distance between two practitioners is taken into account. It was found that such a pair of parameters nevertheless achieved a good geographical spread of the practitioners to whom a wait was assigned in a simple manner.
    It was found that the history of the performed watchings can be kept in the form of a waiting capital in a predetermined period of time. In such an embodiment of the present invention, the history of the watchings performed includes the waiting capital. The time span is, for example, a year. However, this is not necessary for the present invention and also longer periods of time such as, for example, a decade, etc., or shorter periods of time such as, for example, a trimester, semester, season, month, etc. can be used. The waiting capital is accrued by assigning a score to the various waiting sessions or waiting sections. For the waiting sections listed above, for example, a week watch can yield 1 point per day, which means for example 4 points in total per waiting week, a weekend watch can also yield 1 point per day, which means for example 3 points per waiting week, and a night watch also 1 point per night. For public holidays it can be determined that an extra point can be awarded. The number of points allocated to the different waiting times or waiting parts allows the method to assign a certain weight or rating to the different waiting parts, whereby the practitioner can be encouraged to carry out certain waiting times or waiting parts that are otherwise difficult to complete. Although keeping the history of performed waiting times / waiting parts within a predetermined period of time per practitioner can easily be achieved by keeping a waiting capital, this is not necessary for the present invention and other methods can also be used. Although it is possible to determine the individual (minimum) waiting capital per practitioner within the predetermined period of time, it is possible to strive for an average waiting capital per practitioner in the predetermined period of time.
    In preferred embodiments of the present invention, the history of performed queues includes the date of the last performed wait and / or the type of the last performed wait. The type of waiting means, for example, whether the waiting is a public holiday, a night watch, a week watch, a weekend watch, whether or not during the week, etc. This has the advantage that performing specific waiting parts / waiting sessions during a predetermined period can have a specific consequence for the practitioner. For example, it becomes possible to foresee that if a practitioner performs a waiting part / wait of a specific type, for example a public holiday, a specific weekend, a weekend watch, a week watch, a day watch, a night watch and / or combinations thereof such as a night watch over the weekend on a public holiday, etc., the practitioner is exempt from having to be that watch from that specific type of watch during a predetermined period, such as for example a decade or shorter or longer. It also becomes possible to look at determining a minimum and / or maximum time span for the difference in time between two waiting parts / waiting times regardless of the type of waiting time.
    In preferred embodiments of the present invention, the history of performed waiting includes the number of days the practitioners are registered in a register. For example, the register keeps track of how long a practitioner has been a part of the system in which the method of the present invention is applied to control waiting times. This information can then be used, for example, to calculate a so-called waiting frequency on the basis of the accumulated waiting capital by dividing the waiting capital by the number of days that the practitioner is registered in the register that is also kept for this purpose in the history of the performed waiting sessions.
    In preferred embodiments of the method according to the present invention for controlling queuing, the history of the queuing performed includes the time span to the last queuing performed. By including this parameter in the history it can be checked whether a certain pre-set maximum number of days of rest between two waiting times is not exceeded.
    In order to keep an agenda of availability for the waiting sessions per practitioner, the various practitioners can specify a number of free periods, also called black points, in which they do not wish to perform a waiting session. These free periods can, for example, be subdivided into different classes, whereby the periods of different classes are, for example, subject to different predetermined conditions. For example, it is possible to subject the free periods in which the practitioner wishes to take a holiday to a maximum of 6 week and weekend waiting times, including or not including night waiting times. Other classes of free periods, such as for economic reasons, for example renovations, annual fairs, road works, etc. or due to illness, are not subject to a maximum duration.
    Determining the desired number of practitioners who must perform a wait at a certain time within the waiting region, taking into account the population of the waiting region to be served, is preferably done by a coverage number, preferably determined by a minimum, maximum and optimal coverage figure, which indicates how many inhabitants a practitioner is expected to serve in function of the waiting area. For example, it is possible to make the coverage figure dependent on the waiting part and the coverage figure of a day watch is smaller than the coverage figure of a weekend watch which in turn is smaller than the coverage figure of a night watch. Depending on the coverage number within a waiting area and the number of inhabitants of the waiting region, for example represented by the population density, it can then be determined how many practitioners should participate in a waiting session in order to serve all residents within the waiting region. In addition, it is also an advantage that the location of the various practitioners within the waiting region is taken into account in order to achieve an improved spread of the practitioners within the waiting region. For example, the distance, being, for example, the actual walking distance, cycling distance or driving distance, can also be taken into account, partly depending on the time of the waiting section (week, weekend or night watch), between different practitioners. In such a method, the waiting region is preferably as large as possible, since then an optimum spread is achieved and less frequent adjacent waiting regions occur where a reduced good spread may result in the border region.
    According to preferred embodiments of the present invention, the number of possible practitioners for each wait to be performed is determined on the basis of a basic coefficient, the basic coefficient being equal to the waiting frequency of a practitioner on the basis of the accumulated waiting capital and the number of days that the practitioner has already is enrolled in the system and where the basic coefficient is equal to a preset waiting coefficient if the number of days that the practitioner is already enrolled in the system equals zero. It has been found that such a simple parameter still makes it possible to give a weight to the various practitioners in a simple manner when selecting a practitioner who is eligible to perform a specific wait.
    According to further preferred embodiments of the present invention, the waiting frequency thus obtained is scaled down to a waiting coefficient, the scaling taking into account the calculated minimum waiting frequency, maximum waiting frequency and / or average waiting frequency. The average waiting frequency is taken as a waiting coefficient of 1. The maximum waiting frequency gets a waiting coefficient value of 0. This means that the practitioners with a waiting frequency lower than the average are assigned a waiting coefficient that is greater than 1. The practitioners with a waiting frequency higher than 1. the average is assigned a waiting coefficient that is between 0 and 1. In this way, the practitioners who have been on duty the most are assigned the lowest weight (waiting coefficient).
    According to further preferred embodiments of the present invention, the history of the performed wait times includes the time span to the last performed wait. It has been found that such an easily maintained parameter can nevertheless easily be used in the method.
    According to further preferred embodiments of the present invention, the number of possible practitioners for each wait to be performed is determined on the basis of a ramp-up multiplier, wherein the ramp-up multiplier is a number equal to 0 during a practitioner's rest period, the rest period being equal to a pre-set minimum number of rest days between 2 waiting cycles, and the first day after the practitioner's rest period, equal to 1 on the first day after a preset ramp-up period, and increasing from 0 to 1 during the ramp-up period. It has been found that such a simple parameter still makes it possible to give a weight to the various practitioners in a simple manner when selecting a practitioner who is eligible to perform a specific wait.
    According to further preferred embodiments of the present invention, the number of possible practitioners for each wait to be performed is determined on the basis of a too much rest multiplier, the too many rest multiplier being a number calculated by the number of days since the last wait over a pre-settable maximum number of rest days between 2 waiting cycles, multiplied by a pre-settable daily penalty for exceeding the maximum number of rest days and increasing that result by 1. It was found that such a simple parameter still allows to give a weight to the different practitioners in a simple way when selecting a practitioner who is eligible to perform a specific wait.
    According to further preferred embodiments of the present invention, the number of possible practitioners for each wait to be performed is determined on the basis of a region multiplier which, when determining the practitioner, gives a predetermined weight to the geographical environment of the various practitioners. It has been found that such a simple parameter still makes it possible to give a weight to the various practitioners in a simple manner when selecting a practitioner who is eligible to perform a specific wait.
    According to further preferred embodiments of the present invention, determining the number of possible practitioners for each wait to be performed is based on a selection coefficient, the selection coefficient being calculated by raising the basic coefficient to the power determined by a preset square coefficient of basic coefficient and having this result multiply by the ramp-up multiplier, the too much rest multiplier and possibly the region multiplier. It has been found that such a simple parameter still makes it possible to give a weight to the various practitioners in a simple manner when selecting a practitioner who is eligible to perform a specific wait.
    According to further preferred embodiments of the present invention, determining the desired number of practitioners to perform a wait at a given time within the region, taking into account the population of the region to be served, based on a predetermined minimum and maximum population to be served per practitioner at the time. It was found that such simple parameters can still achieve a good geographic spread of the final selected practitioners for a wait.
    According to further preferred embodiments of the present invention, determining the desired number of practitioners to perform a wait at a given time within the region, taking into account the population of the region to be served and the predetermined minimum and maximum population to be served per practitioner at the time, in the second step through the GIS.
    Preferably, the above information is parameterized:
    Per practitioner: - geographic location - free periods per predetermined period - last time the practitioner was on duty on a specific public holiday - last time a practitioner participated in a wait - maximum number of waiting parts per 2 years to spread the watchings over the practitioners better spread
    Per guard region: - population density - coverage number in function of the waiting area, for example in function of week watch, weekend watch, night watch - minimum distance between two practitioners of watch in function of the waiting part, for example in function of week watch, weekend watch, night watch - maximum distance between two practitioners of guard in function of the waiting section, for example in function of week watch, weekend watch, night watch - minimum number of watchmen in the waiting region in function of the waiting section, for example in relation to week watch, weekend watch, night watch - average waiting capital per practitioner per year Further parameters: - minimum time period between two holidays within the waiting period - minimum time period between two waiting cycles - definition of distance (for example walking distance, cycling distance or driving distance or a travel time taking into account public transport and / or walking distance) - maximum number of nights of guard per practitioner in function of it waiting area, for example in function of week watch, weekend watch, night watch The parameters mentioned above can be changed, can be supplemented, can be omitted, etc. for example in function of specific situations. Optionally, additional parameters can also be set.
    For example, the selection of a practitioner for a certain waiting area is preferably made on the basis of the parameters: - time span since previous similar waiting area (for example: time span since last week's watch, night watch or weekend watch) - time span since the previous specific public holiday on which the practitioner of wait was - waiting capital, preferably the difference between the waiting capital and the average waiting capital taken over all practitioners within the waiting region - free periods indicated by the practitioner
    The selection from the previously selected practitioners for performing a waiting part is then preferably based on: - the geographical location of the pharmacy - the population density of the waiting region - the coverage figure for the waiting part to be performed - minimum distance between two practitioners for wait during the waiting part - optimum distance between two practitioners for guard during the waiting part - maximum distance between two practitioners for guard during the waiting part - core areas
    Possibly different waiting parts or even waiting sessions are linked to each other. For example, it is possible that in order to determine who is eligible to perform a certain waiting period or a certain waiting period, only previously selected practitioners are eligible for another waiting period. For example, it is possible that for a night watch only practitioners who are on watch during the day watch prior to the night watch to be filled in are eligible. The further selection is then carried out, for example, by looking at the requested free periods, the time span since the previous night watch performed, the maximum number of nights of watch per practitioner in function of the waiting egg, etc. This has the consequence that for the complete completion of a waiting week first a first series of waiting parts must be entered and, depending on this, other waiting parts. It is also possible to set the week watch and the weekend watch in such a way that they do not follow one another consecutively and alternate with each other.
    By completing the various waiting sections, the waiting sessions are then created.
    After the completion and completion of the different waiting parts and thus the determination of the different waiting times of the practitioners in the waiting region within the predetermined period of time, for example a year (waiting year), overviews can be made of the interpretation of the waiting year. Such an overview can be, for example, in the form of a waiting card that a practitioner can display and on which the practitioner indicates which practitioner is nearby when and on hold. A waiting list can also be drawn up in which a comprehensive overview is given of the waiting of the various practitioners in the waiting region.
    By filling in all waiting parts / waiting times in this way, a distribution of the different waiting parts / waiting times is obtained for a waiting region within a waiting year among the different practitioners within the waiting region.
    Preferably, the possibility is also provided that the practitioners themselves can at least partially indicate their parameters for each waiting year. In this way it becomes possible for the practitioners to indicate their free periods before commencing the distribution of watchings within a coming waiting year. To this end, an application is preferably provided that allows the practitioners to log in via the internet, for example via a website, and thus pass on their free periods.
    Preferably, the computer system allows the practitioner to request what his current waiting capital is, what the average waiting capital is for all practitioners within the waiting region, etc.
    Preferably, the possibility is also provided for two practitioners to change the waiting part or waiting time and that this switching is registered and that the switching is taken into account in a future distribution of the waiting parts. In a further preference, the practitioners also indicate this via an application via the internet, for example via a website.
    It is also possible that the computer system is provided with a control mechanism that checks whether, due to the switch between the two practitioners, no parameters are being exceeded and consequently, for example, a sufficient spread of the practitioners is not achieved. If such a switch is detected, for example, the switch can be refused, a message can be sent to a system administrator, etc. For example, it is possible to provide a control system that only allows waiting switches between a few nearby practitioners, for example depending on the specific waiting section that has been exchanged for is, for example, that only a day watch can be exchanged with the 4 nearest practitioners and a night watch with the 10 nearest practitioners. If a different waiting change is still required, a manual intervention in the system can for instance be provided, for example by an administrator, who still allows the waiting change.
    The invention also relates to a computer program in a computer readable format, comprising software code sections for carrying out the method according to the invention.
    An example of an algorithm for controlling waiting times within a waiting region is given in Figures 1 - 18 which illustrate a specific embodiment of an algorithm according to the present invention. The algorithms outlined herein are specifically provided for pharmacists. However, it is possible to provide this algorithm for any practitioner within the context of the present invention.
    Finally, Figure 19 is also shown by way of example, in which a distribution of an area, more specifically Antwerp, over 6 pharmacies is shown.
    Another example of an algorithm for controlling queues within a waiting region is given in Figures 20 -XXX illustrating an alternative embodiment of an algorithm according to the present invention. The algorithms outlined herein are also specifically provided for pharmacists. However, it is possible to provide this algorithm for any practitioner within the context of the present invention.
    Figure 1 shows an overview of an algorithm in which a so-called waiting week is initialized. Here, in a first step 1, the next non-completed waiting week is calculated on the basis of the date of the last completed wait (called waiting service referred to in the figures). This waiting week that has not yet been completed is then indicated by a year and a week number.
    In a second step 2, the classification of the waiting week that has not yet been classified is then calculated. The calculation of the waiting week classification is shown, for example, in algorithm A shown in Figure 11. The result of this calculation is a start date of the waiting week, an end date of the waiting week and the waiting parts that fall into this waiting week divided by type, in this case weekend or week, and per day part, in this case day or night. In this way, after performing the algorithm according to Figure 1, the entire next non-completed waiting week is divided into the different waiting sections, which constitute waiting sessions that can be divided among the different practitioners.
    In a third step 3, the waiting week in a calendar is then supplemented. When filling in the waiting week in the calendar, holidays that fall in the waiting week are also indicated.
    Subsequently, in a fourth step 4, this information is used to calculate so-called black points for the practitioners on the basis of a black point parameter in which the number of years for which the practitioner is exempted after a public holiday is proposed. More specifically, for the waiting week to be classified for the holidays in this waiting week on the basis of the waiting calendar and the blackpoint parameter per holiday in the waiting week to all practitioners, in this case pharmacists, who have spent the past x years, ie the number of years of the blackpoint parameter, all these holidays have been awarded a blackpoint. This means that certain practitioners on the basis of such a awarded black point will not be eligible for fulfilling a specific waiting period.
    Figure 2 then shows a first step 5 of the method according to the invention in which the practitioners are selected that can be used for the waiting week to be classified on the basis of the black points of the practitioners for the holidays that are in the waiting week to be classified falls, the registrations of practitioners for that particular waiting week, as a result of which they declare their willingness to fill in certain waiting periods in the waiting week to be classified and a minimum number of days of rest between 2 waiting periods. This information is partially extracted from the practitioners 'calendar and blackpoint parameters as well as the history of the practitioners' performed watchings, as previously explained. Also, although not necessary, some additional data may be kept for each practitioner, such as, for example, a unique number for each practitioner, in the case of pharmacists, for example, an apbnr. and possibly, but not necessarily, a regional multiplier that reflects the population density in the neighborhood of the practitioner. Data is further sought from practitioners who are candidates for classifying a waiting week for the waiting week as long as the waiting arrangement for the waiting week has not been approved.
    Figures 3a and 3b show that in a next step a selection coefficient according to Figure 2 is calculated for a searched candidate which is finally assigned to the searched candidates.
    As can be seen in figures 3a and 3b, the selection coefficient is calculated on the basis of: - A ramp-up multiplier 6 as already discussed above. The ramp-up multiplier preferably increases linearly. However, this is not necessary and the ramp-up multiplier can also increase from 0 to 1 in another way, such as exponentially.
    - Too much rest multiplier 7 as already discussed above.
    - A basic coefficient 8 as already discussed above.
    For scaling, for example, a function can be used that takes into account the desired minimum, maximum and average waiting coefficient when converting to the waiting coefficient. Such a function can be linear. This waiting coefficient is then used as the basic coefficient.
    - A pre-set squaring value for the basic coefficient is also used.
    The calculation of the selection coefficient 9 is done as already discussed above by raising the basic coefficient 8 to the power determined by the squared value basic coefficient and multiplying this result by the ramp-up multiplier 6 and the too much rest multiplier 7.
    In a next step 10 the selection coefficient is added to the data of the candidate.
    Figure 4 shows that a preset run sequence is looked up. Certain waiting times are, after all, linked to each other, as already explained above, so it is important to determine which waiting times have to be filled in first and which waiting times can be filled in (further) depending on previously filled in waiting times. Such a run sequence is characterized by certain waiting types for certain parts of the day with a certain order in the collection of run sequences and an algorithm for filling up a list of candidates. It was found that by filling in the queues by using runs in which related queues are grouped, some runs can even be performed totally independently of each other and can even be performed at the same time, for example by using multithreading. Figure 5 shows that for each run sequence of the waiting week to be classified, a search is made for which waiting cycles belong to a specific run sequence.
    Furthermore, figure 5 shows that per non-initialized run sequence of a run sequence a non-initialized run is created. With each run, the occupancy parameters are given that are preset in the system and the parameters required for a GIS (geographic information system or in the Dutch Geographic Information System). In addition, algorithms are provided for determining the number of practitioners to be eliminated per iteration and for solving problems, as shown in Figures 16-18, such as insufficient candidates and insufficient results after the run has been performed.
    Figure 6 then shows that a run set is created for the waiting week to be assigned based on the different runs that have not yet been initialized, the list of candidates that has not yet been drawn up and the waiting week to be classified. A run planner is then initialized based on the run set. The run planner includes a runscheduler, for scheduling the filling of the candidate lists of the different runs, and a runfiller, which fills the candidate list of the run. For this purpose, the run is added per run in the runset to the runscheduler who adds the run to the runfiller unless the run can already be filled based on another run, based on another run, for example waiting sessions that must be performed by the same practitioner after which the list of candidates for the run is filled on the basis of the algorithm for filling up the list of candidates that has not yet been filled.
    The algorithm for filling up the list of candidates that has not yet been filled is illustrated in Figures 12-15.
    As soon as the list of candidates for a run has been filled, an evaluation is made as to whether sufficient candidates have been included in the list of candidates. This is illustrated in Figure 7. If sufficient candidates are included in the candidate list, the status of the run, i.e. the run status, is considered initialized. If there are not enough candidates in the candidate list of a run, an additional algorithm is executed that is shown in figure 16.
    After initializing the different runs, the runset of the different initialized runs is forwarded to a subroutine that is provided to work with the GIS (geographic information system) system. By providing such a structure in which first a complete set of information is built up (the runset) on the basis of which the GIS system can determine the optimal spread, the structure of the runset (s) can be completely separated from the GIS. This is advantageous since the runset (s) will often be prepared by a first instance (for example, an association of practitioners, such as an association of pharmacists) and a second separate instance manages the GIS and thus further processes the runset (s). In this way, a complete GIS can be purchased / licensed by an association of practitioners who use the GIS only sporadically and the GIS can be operated by specialized companies, which is often more cost efficient.
    The application of the GIS is shown in Figure 8. Based on the GIS, a result list is created by means of a GIS tool for each initialized run until all runs have a filled results list.
    Figure 9 shows how it is checked whether sufficient results are present in the different filled runs. If insufficient results are present in a run, an additional algorithm is executed that is shown in figure 17 or 18. If runs are present that need to be filled, based on the run order, based on previously filled runs, then they are then padded based on an additional algorithm shown in Figure 13.
    Figure 10 then shows the algorithm in which the end of filling up the runs is evaluated and in which the waiting arrangement is approved or rejected, where different overviews, such as for example the control cards, are created and / or printed and finally the statistics needed for a following calculation of a waiting arrangement.
    Figure 20 is substantially the same as Figure 1, Figure 28 being very similar to Figure 11.
    Figure 21 is substantially the same as Figure 4. However, Figure 1 connects immediately to Figure 4 without the intermediate steps shown in Figures 2 and 3. Figure 21, however, shows an additional step in which the unfilled waiting weeks are looked up.
    Figure 22 contains many similarities to Figure 5, with the difference that no occupancy parameters or GIS parameters are included, but that only run sequence, wait and wait week are used to create a run.
    Figure 23 shows many similarities with Figure 6, but in the method shown in Figure 23 a candidate list is not taken into account.
    Figure 24 shows that in a next step as long as there are runs with the status initialized, runs are filled using a calculation algorithm C1 shown in Figure 30.
    Fig. 30 shows that on the basis of algorithm B1, shown in Fig. 29, which is very similar to Figs. 3a and 3b, a number of possible practitioners are determined for each wait to be performed account, preferably taking into account the availability and the already performed waiting cycles. in a first step.
    The assignment of the waiting to be performed to at least one of them, being the possible waiting practitioners determined in the first step, taking into account the availability and the already performed waiting, taking into account the geographic spread, is carried out in a subsequent second individual step.
    Figure 30 further shows that this second step is done using some algorithms included in algorithm D1 shown in Figure 32. A first algorithm makes a selection from a series of candidates, taking into account the weight of the candidate practitioner and the practitioners. remove geographically too close to already selected practitioners. Then all practitioners are removed that are in a core area in which at least 1 practitioner has already been selected. In a next step, one practitioner is selected per core area without already selected practitioners. If this core area is adjacent to one or more core areas in which a practitioner has already been selected, the practitioner is selected with optimum distance from the selected practitioners, otherwise the practitioner with the largest weight is selected. Subsequently, all other practitioners are removed in the core area. Again, the practitioners are removed that are too close to the selected practitioners and the practitioners are selected outside the core areas. This is done by assigning the population within the area to the nearest practitioner on the basis of the "shortest path algorithm" and selecting the practitioner who is optimally located relative to the selected practitioner with the most population who does not border on the area of another selected practitioner Finally, those selected practitioners are removed who, upon removal, do not cause the maximum population of another selected practitioner to be exceeded A second algorithm C2 shown in Figure 31 splits the candidates into 2 equally well-spaced series. use is made of graph coloring which is shown in more detail in figure 33 by algorithm D2.
    Figures 25, 26 and 27 correspond to the further settlement similar to the settlement shown in Figure 10 for making further audit reports, etc.
    权利要求:
    Claims (17)
    [1]
    Method for arranging watchings within a group of practitioners of the same occupation for a predetermined region, with the following steps: - keeping a history of performed watchings per practitioner; - keeping an agenda of availability for the waiting sessions per practitioner; - determining the practitioners who must perform a wait at a certain time within the region, taking into account the population of the region to be served, - determining a number of possible practitioners for each wait to be performed and assigning the wait waiting to be performed on at least one of them, taking into account the geographic spread, the availability and the already performed waiting times, the method being carried out at least in part by at least one computer system.
    [2]
    A method for controlling queues according to claim 1, characterized in that determining a number of possible practitioners for each queue to be performed taking into account the availability and the queues already performed is performed in a first step and assigning the Perform waiting on at least one of them, taking into account the geographic spread, is performed in a subsequent second separate step.
    [3]
    Method for controlling waiting times according to claim 2, characterized in that the second step is carried out by means of a GIS.
    [4]
    Method for controlling waiting times according to claim 3, characterized in that the first and the second step are performed by a first and a second computer system, respectively.
    [5]
    Method for controlling waiting times according to claim 3 or 4, characterized in that when assigning the waiting times to be performed to at least one of them, taking into account the geographic spread, a preset minimum and maximum distance is taken into account between two practitioners.
    [6]
    Method for controlling waiting times according to one of the preceding claims, characterized in that the history of performed waiting times contains a waiting capital in a predetermined period of time, the waiting capital being built up by assigning a score to the various executed waiting sessions or waiting parts know.
    [7]
    Method for controlling waiting times according to one of the preceding claims, characterized in that the history of performed waiting times contains the date of the last performed wait and / or the type of the last performed wait.
    [8]
    Method for controlling waiting times according to one of the preceding claims, characterized in that the history of performed waiting times contains the number of days that the practitioners are registered in a register.
    [9]
    9. Waiting control method according to claim 8 in combination with at least claim 6, characterized in that the number of possible beta practitioners for each wait to be performed is determined on the basis of a basic coefficient, the basic coefficient being equal to the waiting frequency of a practitioner on the basis of the accrued waiting capital and the number of days that the practitioner has already been registered in the system and the basic coefficient is equal to a preset waiting coefficient if the number of days that the practitioner has already been registered in the system equals zero.
    [10]
    A method for controlling waiting times according to claim 9, characterized in that the waiting frequency thus obtained is scaled down to a waiting coefficient, whereby the calculated minimum waiting frequency, maximum waiting frequency and / or average waiting frequency are taken into account when scaling.
    [11]
    Method for controlling waiting times according to one of the preceding claims, characterized in that the history of the performed waiting times contains the time span up to the last waiting wait performed.
    [12]
    A method for controlling queuing according to claim 11, characterized in that the number of possible practitioners for each queuing to be performed is based on a ramp-up multiplier (6), wherein the ramp-up multiplier (6) is a is a number equal to 0 during a practitioner's rest period, the rest period being a pre-set minimum number of rest days between 2 waitings, and the first day after the practitioner's rest period, being 1 in the first day after a pre-set ramp-up period, and increases from 0 to 1 during the ramp-up period.
    [13]
    Method for controlling waiting times according to claim 11 or 12, characterized in that the number of possible practitioners for each waiting time to be performed is determined on the basis of a too much rest multiplier (7), the too many rest multiplier is a number that is calculated by multiplying the number of days since the last wait over a predetermined maximum number of days of rest between 2 waiting cycles, by a predetermined daily penalty for exceeding the maximum number of days of rest and calculating that result increase by 1.
    [14]
    Method for controlling waiting times according to one of the preceding claims in combination with at least claims 9, 12 and 13, characterized in that the number of possible practitioners for each waiting turn to be performed is based on a selection coefficient (9) , wherein the selection coefficient is calculated by raising the base coefficient (8) to the power determined by a preset squaring value base coefficient and multiplying this result by the ramp-up multiplier (6) and the over-resting multiplier (7).
    [15]
    Method of controlling waiting times according to one of the preceding claims, characterized in that the determination of the practitioners who must perform a waiting period at a specific time within the region, taking into account the population of the region to be served, takes place based on a minimum and maximum number of population to be served per practitioner at the time.
    [16]
    A method of controlling queuing according to claim 15 in combination with at least claim 3, characterized in that determining the practitioners who must perform a queue at a given time within the region, taking into account the population of the region that is served and the predetermined minimum and maximum number of population to be served per practitioner at the time, is done in the second step by means of the GIS.
    [17]
    A computer program in a computer readable format, comprising software code portions for performing the method of any one of claims 1-16.
    类似技术:
    公开号 | 公开日 | 专利标题
    Fikar et al.2017|Home health care routing and scheduling: A review
    Braekers et al.2016|A bi-objective home care scheduling problem: Analyzing the trade-off between costs and client inconvenience
    Puente et al.2009|Medical doctor rostering problem in a hospital emergency department by means of genetic algorithms
    Rodriguez et al.2015|Staff dimensioning in homecare services with uncertain demands
    Sinreich et al.2012|Reducing emergency department waiting times by adjusting work shifts considering patient visits to multiple care providers
    Shao et al.2012|The therapist routing and scheduling problem
    De Vuyst et al.2014|Computationally efficient evaluation of appointment schedules in health care
    Ben-Tovim et al.2016|Hospital event simulation model: Arrivals to discharge–design, development and application
    Vissers et al.2007|Developing a platform for comparison of hospital admission systems: An illustration
    Nasir et al.2020|A decision support framework for home health care transportation with simultaneous multi-vehicle routing and staff scheduling synchronization
    Jacobson et al.2013|Discrete-event simulation of health care systems
    Elalouf et al.2015|An alternative scheduling approach for improving patient-flow in emergency departments
    Anjomshoa et al.2018|An exact approach for tactical planning and patient selection for elective surgeries
    Morikawa et al.2017|Scheduling appointments for walk-ins
    Gunpinar2013|Supply chain optimization of blood products
    Camiat et al.2021|Productivity-driven physician scheduling in emergency departments
    BE1019491A5|2012-07-03|METHOD OF CONTROLLING WAITING WITHIN A GROUP OF SAME PROFESSIONALS FOR A PREFERRED REGION AND A COMPUTER PROGRAM IN A COMPUTER READABLE FORMAT, INCLUDING SOFTWARE CODE PARTS FOR CARRYING OUT THIS WORK.
    Aroua et al.2015|Forecast emergency room visits–a major diagnostic categories based approach
    Antunes et al.2019|A solution framework based on process mining, optimization, and discrete-event simulation to improve queue performance in an emergency department
    Van Bockstal et al.2019|A study on the impact of prioritising emergency department arrivals on the patient waiting time
    Vandenberghe et al.2021|Stochastic surgery selection and sequencing under dynamic emergency break-ins
    Bai et al.2021|Managing admission and discharge processes in intensive care units
    CN105637484A|2016-06-01|Task assignment method, computer program product and task assignment system
    Radman et al.2018|Designing a multi-service healthcare network based on the impact of patients’ flow among medical services
    Veneklaas2019|Decision support for design, implementation, and feasibility of an admission lounge
    同族专利:
    公开号 | 公开日
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    CA2593371A1|2007-06-25|2008-12-25|Chyma Systems Inc.|System and method for on-call scheduling|
    法律状态:
    2019-01-30| MM| Lapsed because of non-payment of the annual fee|Effective date: 20180531 |
    优先权:
    申请号 | 申请日 | 专利标题
    BE201000630|2010-10-22|
    BE201000630|2010-10-22|
    [返回顶部]