Job cooperation support system and job cooperation support method

An embodiment of the present invention relates to a business cooperation support system and a business cooperation support method that support business cooperation in a field where a plurality of staff members work in cooperation with each other.

Traditionally, in workplaces where multiple employees of multiple occupations work together, such as maintenance services for hospitals, nursing homes, restaurants, and facilities (buildings, elevators, power plants, factories, etc.), cooperation is facilitated and work is performed efficiently. In order to do so, it is necessary to grasp the progress of other work related to the work that you plan to perform. In hospitals, the general method is to use a mobile phone (in-hospital PHS) to make inquiries to the staff who are carrying out related work, but it is impossible to grasp if the inquiring party cannot answer the phone. Is. On the contrary, it is possible to contact the related staff by telephone in order to let others know the progress of the work, but there is also a problem that the other person cannot respond and forgets the act itself.

In addition, a system that can record work execution, such as an electronic medical record, may be able to grasp the progress of related work, but in such a system, it is not possible to grasp the information unless you check it yourself, so the confirmation act itself If you forget, you will not be able to grasp it. In addition, since the work execution record is not always performed immediately after the work is carried out, it may lack real-time performance. In addition, if there is no detailed execution record of the work, there is a problem that the progress of the work cannot be grasped by such a system.

In addition, a system that recognizes work using only IC tags and associates it with the voice confirmation of the worker is known, but in this system, when multiple tasks are performed in the same place such as a hospital. Had the problem of not being able to recognize which work was being carried out. On the other hand, a system that identifies work using only voice data is also known, but in this system, the worker regularly details which work has started, is being carried out, has been completed, and the like. There was a problem of having to speak. In addition, a simple combination of these technologies solves the problem that humans must voluntarily and frequently take detailed actions for recording and occurrence because there is no cooperative operation between the system and the user. Can not.

<p><patcit num="1"><text>Japanese Unexamined Patent Publication No. 2008-186333</text></patcit><patcit num="2"><text>Japanese Unexamined Patent Publication No. 2008-250664</text></patcit></p>

<p> One aspect of the present invention provides a business cooperation support system and a business cooperation support method capable of supporting business cooperation between staff members without imposing a burden on the staff members.</p>

<p> The business cooperation support system as one aspect of the present invention is a business cooperation support system for each staff member who holds a mobile terminal, and is estimated to be a business information holding means, a sensor data receiving means, and a situation holding means. It is provided with a means and a means for generating and delivering a tweet message.</p><p> The business information holding means holds information on the work schedule and work flow of the staff.</p><p> The sensor data receiving means receives sensor information from a mobile terminal held by the staff.</p><p> The status holding means holds the business status including the received sensor information.</p><p> The estimation means calculates the accuracy of the predetermined execution status of the work of the staff based on the information held in the business information holding means and the sensor information, and the staff according to the calculated accuracy of the predetermined execution status. Estimate the work being carried out by the company and its implementation status.</p><p> The tweet message generation / delivery means creates a message according to the business execution status estimated by the estimation means, determines a mobile terminal of an employee to whom the message is delivered, and supplies the determined mobile terminal to the determined mobile terminal. Deliver the message.</p>

<figref num="1">The block diagram which shows the business cooperation support system which concerns on the basic embodiment of this invention.</figref><figref num="2">The block diagram which shows the business cooperation support system which concerns on embodiment of this invention.</figref><figref num="3">The figure which shows the structure of the mobile terminal.</figref><figref num="4">The figure which shows the business schedule information.</figref><figref num="5">The figure which shows the business information table.</figref><figref num="6">The figure which shows the business flow information table.</figref><figref num="7">The flowchart of the operation of the first estimation means.</figref><figref num="8">The figure which shows the relationship between various business schedule ID lists created by the 1st estimation means.</figref><figref num="9">The figure which shows the example of the business implementation status estimation result.</figref><figref num="10">Flowchart of operation of inquiry means.</figref><figref num="11">Flowchart of operation of response receiving means.</figref><figref num="12">The flowchart of the operation of the second estimation means.</figref><figref num="13">Operation flowchart of business progress communication means.</figref><figref num="14">The figure which shows the work schedule and the staying position of a nurse.</figref><figref num="15">The figure which shows the average execution time of work.</figref><figref num="16">The figure which shows the average moving time between two positions.</figref><figref num="17">The figure which shows the work schedule execution order of a nurse.</figref><figref num="18">The figure which shows the flow line length of each of two execution sequences.</figref><figref num="19">The figure which shows the example of the time required when the work is performed without using an embodiment.</figref><figref num="20">The figure which shows another example of the time required when the work is performed without using an embodiment.</figref><figref num="21">The figure which shows the tweet data browsing page.</figref><figref num="22">The figure which shows the tweet list display of the member of a specified department.</figref><figref num="23">The whole block diagram which concerns on the sixth embodiment.</figref><figref num="24">A block diagram of the means for determining the delivery of tweet messages.</figref><figref num="25">The figure which shows the situation estimation result.</figref><figref num="26">The figure which shows the tweet delivery destination attribute selection rule.</figref><figref num="27">The figure which shows the example of the delivery destination attribute table.</figref><figref num="28">A flowchart showing the procedure for determining the delivery of a tweet message.</figref>

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

(Basic embodiment) FIG. 1 shows the configuration of a business collaboration support system according to a basic embodiment of the present invention.

This business cooperation support system includes a sensor data receiving means 51, a business information holding means 52, a status holding means 53, a tweet message generating / delivering means 54, and an estimation means 21. The business collaboration support system uses these means to support business collaboration between each staff member who holds the mobile terminal 11.

The business information holding means 52 holds information on the work schedule and work flow of the staff.

The sensor data receiving means 51 receives the sensor information from the mobile terminal 11 held by the staff.

The status holding means 53 holds the business status including the sensor information received by the sensor data receiving means 51. The estimation means 21 calculates the accuracy of the predetermined execution status of the work of the staff based on the information held in the business information holding means 52 and the sensor information held in the status holding means 53, and the calculated accuracy of the predetermined execution status. According to the above, the work carried out by the staff and the implementation status thereof are estimated. The estimation result of the implementation status is held in the status holding means 53.

The tweet message generation / delivery means 54 creates a message according to the business execution status estimated by the estimation means 21, determines the mobile terminal 11 of the staff to whom the message is delivered, and applies the determined mobile terminal to the determined mobile terminal. Deliver the message.

Hereinafter, the first, second, third, fourth, fifth, and sixth embodiments based on such a basic embodiment will be described. Details of the operation of each part of FIG. 1 will be further clarified in the description of these embodiments.

(First Embodiment) In the embodiment of the present invention, the work being performed by the staff member is estimated based on the sensor information and the work schedule, and the staff member is inquired as to whether or not the estimated work is completed. The purpose is to estimate the business progress without multiplying. Further, in the embodiment of the present invention, by receiving the response of the staff member to the inquiry and re-estimating the work progress status based on the content of the response, the other staff members who need to know that the work has been completed in real time. The purpose is to inform you that the work has been completed.

More specifically, an embodiment of the present invention relates to a business cooperation support system and a business cooperation support method in a field where a plurality of people work in cooperation with each other, having the following functions. 1) For each employee, estimate the current implementation work of the employee from the voice made by the employee, the current time, the current position of the employee, the goods used, the work schedule of the day, and the execution history of each work on the work schedule. function 2) A terminal that each employee holds an inquiry message (voice or text) that prompts each employee to respond to the progress of the work in a timely manner according to the estimated current work of each employee. Ability to send 3) A function to receive a simple response (voice, button press information on the terminal screen, tweet data such as text) issued by each staff member to the terminal in response to the above inquiry message. 4) A function to re-estimate the progress of the work from the received response. Five) A task that makes it easier for other staff to understand the progress of the task by complementing or converting the received response using information about the task, such as the name of the task and the name of the service target (patient, etc.) of the task. Ability to create progress contact messages 6) A function to send a business progress notification message (voice or text) to a terminal held by another staff member who needs to grasp the business progress status of the business. 7) In the estimation of the staff's current work to be carried out in 1 above, if there are multiple work schedules in the same time zone on the work schedule of the day, it cannot be carried out by using the above tweet data. A function to improve the estimation probability of the execution work by excluding the work schedule from the work estimation result.

FIG. 2 is a block diagram showing a business collaboration support system according to the first embodiment of the present invention.

The sensor data receiving means 51 includes a tweet data receiving means 12 and a position data receiving means 17.

The business information holding means 52 includes a business schedule holding means 2 and a business information holding means 3.

The status holding means 53 includes a tweet data holding means 13, a staff position history holding means 4, and a status holding means 15.

The tweet message generation / delivery means 54 includes a response receiving means 7, an inquiry means 6, a business progress communication means 9, and a tweet data viewing means 14.

The estimation means 21 includes a first estimation means 5 and a second estimation means 8.

The business schedule holding means 2 holds business schedule information including a list of business schedules of each employee by day or working hours.

The business information holding means 3 sets a set of business ID, business flow ID, business name, job type, equipment used ID, average execution time, and standard deviation of execution time for each business such as receiving instructions, mixed injection, and performing injection. Holds the business information table including.

In addition, the business information holding means 3 holds a business flow information table including business flow information indicating the preceding and succeeding relationships between the businesses. The work flow includes, for example, an injection flow, an internal medicine flow, and a blood collection flow. Each of the above operations constitutes one procedure in the business flow.

The staff position history holding means 4 holds a history of staff position information. The history of the staff's location information can be collected by the location data receiving means 17 at regular time intervals τ by various means such as RFID to be carried by the staff, Wi-Fi of a smartphone, indoor GPS, and Zigbee. The staff position history holding means 4 internally stores the position information acquired by the position data receiving means 17. The acquisition of position information is not limited to this, and can be acquired by using any sensor means. For example, the position information of the staff may be acquired by analyzing the image acquired by the camera installed in the building. It is also possible to pick up the sound in the building using a microphone and combine the picked up sound with the result of image analysis to narrow down the position of the staff more accurately. It is also possible to acquire location information using GPS.

In addition, the staff position history holding means 4 uses the staff position information history to acquire staff stay start time information indicating when the staff started staying at that position, and holds the staff stay start time information internally. ..

The first estimation means 5 estimates the current execution work schedule of each staff member at regular intervals, and creates a work execution status estimation result. The first estimation means 5 causes the status holding means 15 to hold the created business execution status estimation result. Further, the first estimation means 5 calculates the executed probability (implemented probability) or the in-implemented probability (in-execution probability) of the work related to the implemented work schedule. In this embodiment, the probability is used as the probability, but another means may be used.

Further, the first estimation means 5 creates a subsequent work schedule ID list, which is a list of business schedule IDs that are premised on the completion of the work related to the work schedule for the estimated current execution work schedule. The first estimation means 5 causes the status holding means 15 to hold the created subsequent work schedule ID list.

The inquiry means 6 includes an inquiry message information holding means, an inquiry message creating means, an inquiry message sending means, and an inquiry message sending time calculating means. The inquiry means 6 performs the following processing using these means. That is, the inquiry means 6 creates an inquiry message inquiring whether the execution of the work schedule specified in the work execution status estimation result has been completed. Then, the transmission time of the inquiry message is calculated, the staff member who is the destination of the inquiry message is specified, and the inquiry message is transmitted to the mobile terminal 11 or the like held by the staff member at the calculated transmission time.

The response receiving means 7 receives a response from the mobile terminal 11 or the like held by the staff to the inquiry message. When the response receiving means 7 receives the response, the response receiving means 7 identifies which inquiry message information corresponds to the response by using the response receiving time and the response transmitting staff ID that can be identified by the mobile terminal ID of the response transmitting source. Then, the inquiry message and the response are associated with each other.

The second estimation means 8 uses the content of the response from the staff to the inquiry message associated with the response receiving means 7, and estimates the execution work schedule ID of the inquiry message information. And update the in-probability.

The work progress communication means 9 selects a work schedule having a probability of being carried out of 1 or a certain value or more from the work schedules specified by the work execution status estimation result. Then, a message is created to inform that the execution of the work schedule has been completed, and the message is transmitted to the mobile terminal 11 held by the staff in charge of carrying out the work schedule for the work schedule specified in the subsequent work schedule list. ..

The tweet data receiving means 12 receives the tweet data transmitted from the mobile terminal 11 held by the staff.

As shown in FIG. 3, the mobile terminal 11 held by the staff has a data input means 15, and receives the voice emitted by the staff as data. The data input means 15 may accept input of text data and image data in addition to voice data. These data received by the data input means 15 are referred to as tweet data. The tweet data transmitting means 16 transmits the tweet data together with accompanying information such as the current time, the current position of the staff, the staff ID, and the closest patient ID to the tweet data receiving means 12.

There can be various communication media such as LAN (wired or wireless) and 3G between the tweet data transmitting means 16 and the tweet data receiving means 12. The tweet data received by the tweet data receiving means 12 and its accompanying information are held by the tweet data holding means 13.

The first estimation means 5 can improve the estimation accuracy of the execution work by using the tweet data held in the tweet data holding means 13.

Specifically, when there are a plurality of business schedules in the same time zone on the business schedule of the day, the business schedule that cannot be implemented can be excluded from the business estimation results by using the above tweet data. As a result, the estimation probability of the execution work can be improved, and the estimation of the execution work can be performed with higher accuracy.

For example, suppose that between 10 am and 11 am, nurse A has two work schedules to measure vital data (body temperature, blood pressure, etc.) of patient B and patient C. Assuming that the closest patient ID in the tweet data is that of patient B at 10:05 am, the task of measuring the vital data of patient C at 10:05 am is It can be determined that it has not been implemented and can be excluded from the business estimation results. From this, it can be determined that there is a high probability that the work of measuring the vital data of patient B is being carried out.

Here, the estimation execution work may be added to the tweet data held in the tweet data holding means 13.

Further, when the tweet data is voice data, the voice data can be converted into text by a voice recognition means and added as voice text data.

Further, from this voice text data, only words with appropriate part of speech can be extracted by morphological analysis, and important vocabulary can be extracted from voice text data by keyword matching.

The tweet data viewing means 14 displays the tweet data (including the above-mentioned voice text data and keywords in the voice text data) held in the tweet data holding means 13. The tweet data viewing means 14 can also narrow down the tweet data to be displayed according to the attributes (department, occupation, qualification, etc.) of the user who browses the tweet data and the staff ID included in the tweet data.

(Business schedule holding means 2) Figure 4 shows an example of the business schedule information held by the business schedule holding means 2.

Each staff member has multiple work schedules during a certain time period such as AM8: 00 to AM10: 00. The actual work order is decided by the staff according to the situation.

Each data of business schedule information is called business schedule. The work schedule is based on the work schedule ID, staff ID, work ID, patient ID for work, execution position ID, earliest time, latest time, and completed probability (0: definitely not performed, 1: definitely completed). Become.

Here, the earliest time is the time at the beginning of the time zone in which the work specified in the work schedule is carried out, and the latest time is the time at the end of the time zone in which the work schedule is carried out.

In addition to the above-mentioned work schedule information, the status holding means 15 holds the work execution status estimation result, the preceding work schedule ID list, and the succeeding work schedule ID list created by the first estimation means 5.

The "0001" on the left side of the work schedule ID "0001-001" is the staff ID, and the "001" on the right side is the ID of the scheduled work. "0001" on the left side of the business ID "0001-001" is a business flow ID, and "001" on the right side is a task ID.

(Business information retention means 3) As described above, the business information holding means 3 holds the business information table and the business flow information table.

Figure 5 shows an example of a business information table.

Figure 6 shows an example of the business flow information table. In the example of FIG. 6, two examples of business flow information are shown. The business flow information holds a business flow ID, a business ID, a preceding business ID, and a succeeding business ID for each business.

Each rectangle in FIG. 6 indicates a business, and the character strings (1-1, 1-2, etc.) in the rectangle indicate a business ID. These operations are connected by arrows indicating the execution order to form a business flow.

The preceding work is a work to be carried out immediately before the work and generally has a plurality of work. Subsequent work is a work to be carried out immediately after the work, and generally there are a plurality of work. It is possible that the preceding business and the succeeding business do not belong to the same business flow.

In FIG. 6, the succeeding tasks of the task 1-1 are 1-2 and 1-3, and the preceding tasks are empty. The preceding operations of operations 1-5 are 1-4 and 2-3, and the succeeding operations are empty.

(Staff position history retention means 4) [Staff location history information] The staff location history information is a history of staff location information. The staff's position information is acquired from the staff's mobile terminal by the position data receiving means 17 at regular time intervals τ. RFID may be mounted on the mobile terminal. The position data receiving means 17 may receive data via an antenna installed in the building.

The staff position history information is a set of observation time, staff ID, and position ID, and is held in the staff position history holding means 4. The staff ID may be obtained from the RFID if the staff ID is written in the RFID, or can be obtained from the login information of the application of the mobile terminal.

[Staff stay start time information] At the current time t, if the position ID of a certain employee A indicates a certain place p over all the observation times from a certain time ts before t to the present time t, the employee A is after ts, It can be estimated that the user is at the position s until the current time t.

In this case, the time ts is set as the stay start time of the position s of the staff A at the time t. This stay start time will be updated when the current position of the staff changes.

The set of the staff ID, the position ID, and the stay start time is calculated and held by the staff position history holding means 4 as the staff stay start time information.

(First estimation means) A flowchart of the operation of the first estimation means 5 is shown in FIG. Each step of the flowchart will be described in detail below.

[Creation of business implementation status estimation result: S101] As shown in FIG. 8, the first estimation means 5 refers to the staff stay start time information at the current time t and obtains the stay position s of each staff.

Subsequently, the first estimation means 5 selects the estimated execution work of the employee at the current time t from the work schedule information of the employee. Specifically, the work in which the time t is included between the earliest time and the latest time, the probability of execution is 0 (that is, definitely not executed) or a certain value p or less, and the execution position is s. Is selected as the estimated implementation work of the staff concerned.

The first estimation means 5 adds the work schedule ID of the selected work to the estimated execution work schedule ID list. The status holding means 15 holds the estimated execution work schedule ID list.

Here, if there are K estimated execution work schedules at time t of a certain employee, the execution probability p of each estimated execution work schedule k (= 1, 2, 3, ..., K)_kCan be calculated by a multi-nominal logistic regression model (Multi-Nominal Logistic Regresstion) such as the following equation.<img file="JP2012226449A_D0001.tif" /> Here, the variables included in x mean the following, and it is not always necessary to use all the variables. α is a vector having the same dimension as x, and the value of each element of α is calculated in advance by a known method. The acceleration may be obtained by image analysis, or the value of the acceleration clock attached to the staff may be obtained from the mobile terminal.

<img file="JP2012226449A_D0002.tif" /> voice (t) is a word vector composed as follows. Words are extracted by morphological analysis from the voice recognition results of the staff concerned within a certain period of time centered on time t. Have a set of words used in the department to which the user belongs in advance. Arrange the words in lexicographic order and number them starting with 1 as follows. {w1, w2, w3, w4, w5, ..., wN} = {heat, body temperature, pulse, blood pressure, intravenous drip, oral administration, ... patient} Here, if there is a word set extracted from the utterance of the staff concerned within a certain period of time centered on t that matches the above wi, the i-th component is set to 1, and if there is no matching word set, the i-th component is the first. By setting the i component to 0, the following word vector voice (t) can be constructed. voice (t) = (1, 0, 1, 1, 0, 0, 0, ..., 0)

For example, if the voice recognition result of the utterance of the staff member is "Measure heat, pulse and blood pressure", the word "heat" contained in this utterance matches w1 and "pulse" matches w3. , "Blood pressure" corresponds to w4, so the first component, the third component, and the fourth component are 1, so the above word vector can be obtained. The above dictionary can be switched according to the user attributes such as the user's occupation (doctor, nurse, pharmacist) and the department (surgery, internal medicine) to which the user belongs. In addition, words related to the content may be grouped to form a vector having the group as a component. In particular, g1 = {heat, body temperature, pulse, blood pressure, blood oxygen saturation} (words related to vital measurement) g2 = {drip, injection, ...} (words related to injection / drip) g3 = {meal, breakfast, lunch, supper, serving, lower set, ...} (words about meals) g4 ... Then, if there is a word set that matches the above gi, the i-th component is set to 1, and if there is no matching word set, the i-th component is set to 0, so that the vector can be constructed. Alternatively, the ratio of how many of all the words in gi match the words in the utterance may be the i-th component. For example, in the case of "Measure heat, pulse and blood pressure" above, the three words "heat", "pulse" and "blood pressure" match the words in g1. Since g1 has 5 words in total, 3/5 = 0.6 should be the first component of the word vector.

In addition to the multi-value logistic regression model, the Bayesian network, hidden Markov model, etc. can also be used to estimate the probability of completion of each estimated execution work schedule.

In addition to the completed probability, the ongoing probability (probability of execution) q indicates whether or not each estimated execution work schedule is being implemented._kCan be calculated in exactly the same way, and messages can be created and delivered to other staff using the magnitude of the ongoing probability. For example, the in-probability q of an estimated work schedule k_kHowever, if it suddenly increases at a certain time and exceeds a certain threshold value (for example, 95%), it can be estimated that the estimated execution work schedule k has started at that time, so that the estimated execution work schedule k is set by the staff concerned. You can create a message to the effect that it has started and deliver it to other staff to the extent you need it. Other staff members within the necessary range may be, for example, staff members of subsequent work, managers of the staff members, and the like.

Figure 9 shows an example of the work implementation status estimation result obtained by the first estimation means 5.

[Creation of prior work schedule ID list: S102] The first estimation means 5 is, at time t, among the business schedules (data in the business schedule information) specified by the business schedule ID in the estimated execution business schedule ID list, the succeeding business ID with respect to the business schedule ID. The work schedule in which is present is selected as the preceding work schedule at time t.

Whether or not the successor business ID exists can be determined by referring to the data of the business schedule in the business flow information table of the business information holding means 3. The selected business schedule IDs of these prior business schedules are added to the prior business schedule ID list. The status holding means 15 holds the preceding work schedule ID list.

[Creation of business schedule ID list dependent on other business: S103] The first estimation means 5 is a work schedule (data in the work schedule information) in which the start time of all the staff's work schedules is between the current time t and t + Δt after a certain period of time at time t. ), The business schedule for which the preceding business ID exists for the business ID of the business is selected as the business schedule dependent on other business at time t. Whether or not the preceding business ID exists is confirmed in the business flow information table of the business information holding means 3.

Add the selected business schedule IDs for other business-dependent business schedules to the other business-dependent business schedule ID list. The status holding means 15 holds the work schedule ID list dependent on other work.

[Creation of follow-up business schedule ID list: S104] The first estimation means 5 is, at time t, the business schedule A (data in the business schedule information) specified by the business schedule ID in the preceding business schedule ID list and the other business dependent business schedule ID list. About business schedule B (data in business schedule information) specified by business schedule ID (1) Target patient ID of work schedule A = Target patient ID of work schedule B (2) Subsequent business ID of business schedule A = business ID of business schedule B Verify whether both of the above are true.

Instead of (2), (3) the preceding business ID of business schedule B = the business ID of business schedule A may be verified.

If it is established, the business schedule A is in progress (because the business schedule in the preceding business schedule ID list is a part of the estimated implementation business schedule), and the business scheduled to start within a certain time Δt from the current time t. It can be estimated that Schedule B depends on the completion of Business Schedule A.

The first estimation means 5 adds a set of a business schedule ID in the preceding business schedule ID list and a business schedule ID of another business-dependent business schedule that depends on the business to the succeeding business schedule ID list. The status holding means 15 holds the succeeding business schedule ID list.

It is not essential to create the above-mentioned prior business schedule ID list, subsequent business schedule ID list, and other business-dependent business schedule ID list. The business that precedes or follows a certain business schedule may be calculated each time it is necessary by using the business flow information.

(Inquiry method 6) The inquiry means 6 operates according to the flowchart shown in FIG. Each step of the flowchart will be described in detail below. As described above, the inquiry means 6 includes an inquiry message information holding means, an inquiry message creating means, an inquiry message transmitting time calculating means, and an inquiry message transmitting means.

[Inquiry message information retention method] The inquiry message information holding means holds the inquiry message information.

The inquiry message information consists of an inquiry message ID, an inquiry message, an estimated work schedule ID, an inquiry message transmission time, a response, and a sent flag (0: not sent, 1: sent).

The inquiry message is text data, voice data, image data, or the like.

[Compose inquiry message: S201] The inquiry message creating means creates an inquiry message according to the business content of the corresponding business schedule when a new business schedule ID is newly added to the estimated execution business schedule ID list by the first estimation means 5 at the current time t. To do. In addition, an inquiry message ID is assigned to the inquiry message.

The inquiry message creating means creates inquiry message information including the inquiry message and the business schedule ID (estimated execution business schedule ID) of the corresponding business schedule, and sends the inquiry message information to the inquiry message information holding means. The inquiry message means holds this. At this point, the inquiry message does not yet describe the inquiry message transmission time, response, and sent flag.

The inquiry message may be, for example, as follows. That is, let a be the business flow name of the estimated implementation work schedule A and b be the business name. At this time, the inquiry message: "Is b of a finished?"

Here, the business schedule for which the inquiry message creating means creates the inquiry message is the business schedule in which the succeeding business schedule exists among the estimated execution business schedules at time t, that is, the business in the preceding business schedule list at time t. Only the business schedule specified by the schedule ID may be used.

[Calculation of inquiry message transmission time: S202] The inquiry message transmission time calculation means calculates the inquiry message transmission time for each inquiry message information, and sets the time in each inquiry message information.

Specifically, the inquiry message transmission time calculation means first refers to the staff stay start time information held in the staff position history holding means 4, and obtains the stay start time ts of the current position s.

Next, the inquiry message transmission time calculation means uses the business information table of the business information from the business ID of the business schedule specified by the estimated execution business schedule ID of the inquiry message information, and uses the business information table of the business information to obtain the average execution time T and the standard deviation of the business. Inquiry message transmission time: ts + T + Nσ (N = 1,2,3 ...) is calculated with reference to σ. Here, N may be set to an appropriate natural number.

If the staff member has only one estimated work schedule, it can be estimated that the work was started at time ts at position s. Assuming that the distribution of work execution time is normally distributed, statistically, the probability that the work is completed before T + σ is about 84%, and the probability that the work is completed before T + 2σ is about. It can be considered as 98%. If the estimated number of tasks is k (k 2), the inquiry will be made k times.

[Send inquiry message: S203] The inquiry message sending means has a sent flag of 0 in each data of the inquiry message information every fixed time τ, and the current time and the inquiry message sending time are equal (or the difference is less than a certain time). Select things. Then, the selected inquiry message information is transmitted to the mobile terminal held by the staff member specified by the work schedule staff ID specified by the estimated execution work schedule ID. Message information is transmitted in the form of voice messages, text messages or image data. When the transmission is completed, the transmitted flag of the inquiry message information is set to 1.

(Response receiving means 7) The response receiving means 7 operates according to the flowchart of FIG. Each step of the flowchart will be described in detail below.

[Receive response: S301] Correspond the received response with the response reception time. When the response transmission time is assigned to the response on the mobile terminal 11 of the response transmission source, the response transmission time may be used instead of the response reception time below.

[Extraction of staff ID to send recent inquiry message information: S302] The response receiving means 7 refers to the work schedule by using the estimated execution work schedule ID included in the inquiry message information, and obtains the staff ID of the inquiry message transmission destination. At this time, among the inquiry message information, the inquiry message information transmitted when the inquiry message transmission time is a certain time before the response reception time may be selected, and only the selected inquiry message information may be targeted.

[Setting response to inquiry message: S303] Next, the response transmission staff ID that can be identified by the mobile terminal ID of the response transmission source is collated with the staff ID of the inquiry message transmission destination, the inquiry message information is associated with the response, and the content of this response is the response of the corresponding inquiry message information. Set as data.

The response data may be text data, voice data or image data. The text data may be a comment written directly by the staff member, or may be data indicating the content of the staff member's action (click, touch) with respect to the yes / no button displayed on the mobile terminal.

In the above, the response receiving means 7 shows a form in which the response to the inquiry message is received, but the response receiving means 7 sends a voluntary message of the staff (such as Patient A's is over). It may be received from the staff's mobile terminal 11. In this case, the processing of steps S302 and S303 is not necessary.

(Operation of the second estimation means) The second estimation means 8 operates according to the flowchart of FIG. Each step of the flowchart will be described in detail below.

[Selection of response received inquiry message information: S401] The second estimation means 8 selects the inquiry message information for which a response is set and the probability of execution of the business schedule specified by the estimated execution business schedule ID is 0 (or a certain value p or less). Response Received Select as inquiry message information.

[Update of Probability of Execution of Estimated Execution Work Schedule: S402] The second estimation means 8 estimates the executed probability and the executing probability of the business schedule specified by the estimated execution business schedule ID of the selected response received inquiry message information according to the content of the response. Then, when the estimation result is larger than the execution probability of the work schedule at that time, the execution probability and the execution probability of the work schedule of the work execution status estimation result are updated. Furthermore, the completed probability in the work schedule is updated with the same value.

For example, if the response from employee A to the above inquiry message to employee A: "Have you finished b in a?" Is "yes" or "finished", employee A is in job flow a. Is estimated to have been completed, and the probability of completion is 1.

On the other hand, if the response is "No", "Not yet", "It will take a while", etc., it is estimated that Employee A has not completed work a, and the probability of completion is set to 0. Depending on the content of the response, the executed probability may be calculated as a value in the range of 0 to 1.

The probability of implementation was estimated according to the content of the response to the inquiry, but when the response receiving means 7 receives the voluntary message of the staff instead of the response, it is based on the content of this voluntary message. , The same processing may be performed.

Similarly, the probability of execution can be calculated and updated according to the content of the response.

[Update sent flag of inquiry message: S403, S404] If the executed probability is 0 or a certain value p or less, the second estimation means 8 returns the transmitted flag of the corresponding inquiry message information to 0, and sets the inquiry message transmission time to a certain time from the current time t. It may be reset to the time t + Δt after Δt. As a result, the inquiry message is sent again to the mobile terminal 11 held by the staff in charge of carrying out the work schedule at time t + Δt.

(Business progress communication means 9) The business progress communication means 9 is a business schedule specified by the business schedule ID in the estimated execution business schedule ID list (or the preceding business schedule ID list), and the probability of completion is 1 or a certain value or more. Select and create a message to inform that the work schedule has been completed.

The business progress status communication means 9 is a mobile terminal held by the staff in charge of the business schedule specified by the business schedule ID in the subsequent business schedule ID list, in which the created message is specified by the staff member in charge of the business schedule. Send to 11.

The business progress status communication means 9 includes a business progress status communication message information holding means, a business progress status communication message creating means, and a business progress status communication message transmitting means. Figure 13 shows the operation flow of the business progress communication means 9.

[Business progress communication message information retention means] The business progress status contact message information consists of a business progress status contact message ID, a business schedule ID, a business progress status contact message (described later), and a sent flag (0: not sent, 1: sent).

[Create business progress notification message: S501, S502] The work progress communication means 9 has a completed probability of 1 (or a certain value p or more) estimated by the first estimation means 5, or a certain work schedule has a completed probability of 1 by the second estimation means 8. When it is updated to (or a certain value p or more), a business progress notification message is created to inform that the business schedule has been completed.

The business progress notification message may be, for example, as follows. That is, let a be the business flow name of the business schedule and b be the business name. At this time, the business progress notification message: "a b is over" can be used. The business progress communication message may be in any form of text data, voice data, and image data.

The business progress status communication message creation means includes this business progress status communication message and creates business progress status communication message information with the sent flag set to 0. The created business progress status communication message information is held in the business progress status communication message information holding means.

[Send business progress notification message: S503] The business progress status communication message transmitting means identifies the business progress status communication message information held in the business progress status communication message information holding means having a transmitted flag of 0. Then, for the business schedule of the specified business progress communication message information, the subsequent business schedule list is referred to, and for the business schedule specified by the subsequent scheduled business ID, the staff specified by the staff ID in charge of the business schedule is retained. A business progress notification message is sent to the mobile terminal 11.

The business progress communication message may be in any form of text data, voice data, and image data. The business progress communication message sending means sets the sent flag to 1 when the transmission is completed.

Before sending the business progress notification message, the content of the message may be repeated to the staff member, and the message may be sent to other staff members only when the staff member responds with consent.

The data of the message may be stored in the tweet data holding means 13 in association with the ID of the employee and the current time, and may be viewed by the tweet data viewing means 14.

(Second embodiment) In this embodiment, the creation of an estimated execution work schedule ID list using information other than the stay position will be described.

In some cases, such as when a patient is wearing an RFID or an RFID is installed in each patient's bed, it may be possible for staff to obtain information about which patient is approaching at each time.

In addition, when RFID is installed in each medical device, it may be possible for the staff to obtain information on which medical device is approaching at each time.

In addition, since a certain medical device may specify the staff ID at the time of use, in this case, the usage history of the staff for each time can be acquired for each medical device.

In such a case, the first estimation is made by creating the above-mentioned estimation execution work schedule ID list and holding the patient approach information history, the medical device approach information history, and the medical device usage history together with the staff position information history. The means 5 can narrow down the estimated execution work schedule more accurately by using these historical information.

(Third embodiment) [Calculation of inquiry message transmission time according to the estimated execution order of subsequent work schedule] In calculating the inquiry message transmission time described above, the following method may be used. That is, it is estimated that each employee will carry out the work most efficiently for the work schedule of all the staff from the current time t to the time t + Δt after a certain time Δt, and the transmission time of the inquiry message is determined.

Assuming that the execution time of each work alone does not depend on the work execution order and that all the preceding work is completed, the most efficient work execution order is the work so that the movement distance from the position s is the shortest. Is the order in which The first estimation means 5 obtains such a work execution order as an estimated execution order during the time zone [t, t + Δt] at time t.

However, if multiple work schedules for a certain staff member in the above time zone [t, t + Δt] are carried out at the same position, the order of execution for those work is arbitrary.

The estimated implementation order of the time zone [t, t + Δt] of staff A is g1, g2, ..., g (i-1), gi, ..., gn, and the implementation time of each scheduled work gi is , Mean Ti, and standard deviation σi. Based on this estimated implementation order, the sum of the implementation times of the scheduled work g1, g2, ..., g (i-1) is<img file="JP2012226449A_D0003.tif" />Follow the normal distribution of. Therefore, under this estimated execution order, it is considered that the business gi is started after the business schedule g1, g2, ..., g (i-1) is completed. Therefore, the business gi is on average.<img file="JP2012226449A_D0004.tif" />It can be estimated that it will start at.

Here, s (i-1) is the execution position of the scheduled work g (i-1), si is the execution position of the scheduled work gi, and d (s (i-1), s) is the position s (i-1). It is the average moving time required from) to si, and this information can be obtained from the inter-position average moving time information held by the location information holding means in the building, which will be described later.

As a result, the start time of the succeeding business schedule can be estimated for each preceding business schedule. Therefore, the time before a certain time before the estimated start time may be set as the inquiry message transmission time of the preceding business schedule.

[Means for retaining location information in the building] As described above, when the first estimation means 5 estimates the work execution order and uses the average moving time required between the work execution positions, the following in-building position information and the average moving time time information between positions are used. Refer to.

[Location information in the building] For a specific location in the building (such as a hospital room, examination room, nurse center on each floor, etc.), the name and location ID of that location are retained.

[Average time required for moving between positions] For any pair of positions in the building, hold the distance between them or the average travel time divided by the average travel speed.

(Fourth Embodiment) The above-mentioned inquiry message may be transmitted as follows. That is, the inquiry message transmitting means in the inquiry means 6 selects the data whose transmitted flag is 0 from each data of the inquiry message information at regular intervals of τ.

Then, for the work schedule specified by the estimated execution work schedule ID of the selected inquiry message information, the current position ID of the staff is acquired by using the staff ID and the staff position history information.

If the acquired current position ID and the execution position ID of the above work schedule are different, it is determined that the employee has already completed the above work schedule and moved to another position, and the inquiry message is sent to the employee's form terminal. Send to 11.

If the acquired current position ID and the execution position ID of the work schedule are the same, the staff member determines that the work schedule has not been completed yet and waits for the transmission of the inquiry message.

(Concrete example) As shown in Fig. 14, it is assumed that two nurses are performing work in parallel, and each has multiple unexecuted work schedules. It is assumed that the nurse 1 is staying at the place a at the time t, and the work schedules A, B, and C scheduled by the time t + Δt have not been performed. It is assumed that the nurse 2 is at the place d at the time, and the work schedules D, E, and F scheduled by the time t + Δt have not been performed.

Here, it is assumed that B of the work schedules of nurse 1 is premised on the completion of the work schedule D of nurse 2. In addition, it is assumed that the average execution time of each work gA to gE specified by the work schedules A to E is as shown in Fig. 15. It is assumed that the average moving time between the two positions is as shown in Fig. 16.

There are orders 1 to 6 shown in FIG. 17 as the work schedule execution order of nurse 1, and the flow line length of order 1 is the shortest. Assuming that the execution time of a single task is constant regardless of the execution order of the tasks, and that task D by nurse 2 has already been implemented when task B is implemented, all unexecuted task schedules are the shortest. Order 1 is completed. Figure 18 shows the flow line lengths of sequences 1 and 2 and the scheduled work completion time.

However, for example, if nurse 2 performs the work schedule E D at position e in the order of work schedule E D, it takes 4 minutes + 2 minutes = 6 minutes on average, so work D is not completed until time 6 on average. In this case, if nurse 1 executes the work schedule in order 1, as shown in FIG. 19, the work B must be waited for 3 minutes until the time t + 6 minutes, and the work of nurse 1 is performed. The completion time will be t + 14 minutes, which will be 3 minutes later, resulting in inefficiency.

In addition, as shown in FIG. 20, it was found that the work schedule D was not completed after the nurse 1 moved to the position b after the work schedule A was completed, and then moved to the position c to perform the work schedule C. If it is carried out and then returned to b to carry out work schedule B, the work completion time of nurse 1 will be t + 13 minutes and the flow line length will be 5, resulting in inefficiency.

In such a case, using the system of the embodiment of the present invention, if the nurse 1 knows that the task D is completed at the time t + 2 when the task A is completed, the most efficient order 1 Select. If there is no notification that the business D has been completed, it is determined that the business D has not been completed, and by selecting the order 2 in FIG. 18, the business can be completed at time t + 12 minutes. Therefore, as compared with the above two cases (FIGS. 19 and 20), the work schedule can be carried out more efficiently in terms of work execution time and flow line length.

(Fifth embodiment: voice tweet viewing function) When the tweet data browsing page (Web, etc.) as shown in Fig. 21 is opened from the employee's mobile terminal 11 or PC via the tweet data browsing means 14, the server DB (database) (tweet data holding means) is displayed on the page. The tweet data registered in 13) is displayed. When opening a browsing page (Web, etc.), log in by specifying the employee ID.

The displayed contents are -User name: The user name on the user master that corresponds to the user ID of the tweet data. Date: Year / Month / Day Hour: Minute -Location name: The location name on the location master that corresponds to the location ID of the tweet data. -Estimated task name: The task name of the schedule on the schedule data specified by the estimated schedule ID of the tweet data. Keywords: Keywords for tweet data Free memo: Free memo of tweet data -Voice data icon: Click to play the voice data of the tweet data. -Still image thumbnail: The image data of the tweet data is displayed. -Video thumbnail: The video data of the tweet data is played.

On the tweet data browsing page, tweet data can be switched and displayed from various viewpoints such as 1 to 5 below. 1. Menu time series display (as shown in Fig. 21) The default viewing format common to all logged-in users. Like Twitter, it is displayed in chronological order according to the date and time of the tweet data. Regarding the date and time of tweet data, you can narrow down the data to be displayed by the day (within 24 hours), within one week, within one month, all. The default is the day (within 24 hours). Similarly, you can narrow down the date and time of tweet data below. 2. Display a list of tweets of members of the specified department You can select any department in the department master. When selected, attribute 1-ID of the user master displays the tweet data of the user corresponding to the selected department in the descending order of time in the vertical direction as shown in FIG. 22, and displays the users side by side in the column direction. 3. List of tweets about the specified service target person (patient in this case) From the tweet data, the tweet data whose estimated service target person ID belongs to the specified service target person is extracted and displayed in descending order of time in the vertical direction. 4. List of tweets of other users (plural) that the logged-in user wants to follow Show tweets by other users in the follow list for each logged-in user. The display method is the same as in 2 above, the users are arranged in the horizontal direction, and the users are displayed in the descending order of time in the vertical direction. It is necessary to be able to create and save a follow list for each logged-in user. The follow list may be a table on the DB or a CSV file. Five. Display a list of tweets about tasks related to tasks scheduled for the day of the logged-in user For each schedule data of the day of the logged-in user, the tweet data related to the schedule data is extracted (the extraction method is as follows), and the task name of the schedule data is horizontally arranged (arranged from left to right in order of scheduled start time of the schedule). , Display vertically in descending order of time. Here, assuming that certain schedule data of the day of the logged-in user is x, the tweet data related to x is extracted as follows. select * from tweet data table where estimated task ID in (select predecessor task ID from task master where task ID = x. task ID) AND estimated service target person ID = x. service target person ID

[Structure of master, etc.] The master etc. is a table or CSV on the DB.

User master User ID, user name, attribute 1-ID (occupation type), attribute 2-ID (department) * If necessary, have the following file path to the user face photo file as a field. By fixing the storage folder of the face photo file and including the user ID in the face photo file name, the user master may not have the file path to the user face photo file. * For attribute 1-ID, specify the ID defined in the job type master below. * For attribute 2-ID, specify the ID defined in the department master below. User face photo data (PNG files, etc.) Job type master Job type ID, job type name Service target person master Service target person ID, service target person name Department master Department ID, department name Position master Location ID (wireless LAN access point ID), location name -Task master (key: task ID) Task ID, task name, business ID, preceding task ID, succeeding task ID Business master Business ID, business name Schedule data (key: schedule ID) Schedule ID, user ID, task ID, service target person ID, scheduled start date and time, scheduled end date and time, execution position ID, advance schedule ID, subsequent schedule ID

(Sixth Embodiment) The present embodiment has a major feature in the configuration of the tweet message generation / delivery means 54 of FIG. FIG. 23 shows a detailed configuration of the tweet message generation / delivery means 54 in the present embodiment.

The first estimation means 5 (and the second estimation means 8, where the second estimation means is not an essential requirement) starts with sensor information (position information, acceleration information, image information), business information, and tweet information. Create the estimation result 91. The sensor information and the tweet information are received by the sensor data receiving means 51 of FIG. 1 and held by the status holding means 53, and the business information is held by the business information holding means 52 of FIG.

The tweet message generation means 92 generates the tweet message body 951 from the situation estimation result 91 and the tweet information. For example, from the tweet information "Started", add the necessary information using the situation estimation result (location: endoscopy room, patient: A), and "in the endoscopy room, patient A's examination is performed. Generates a tweet message saying "It has started" (information supplement function similar to the business progress communication means 9 of the first embodiment). Note that the tweet information may be arbitrary information such as not only the completion and start of the task, but also the progress (delay, advance) of the task, the notification of awareness, the notification of the message, and the like.

In parallel with this, the tweet message delivery destination determination means 93 generates a delivery destination list 952 from the status estimation result 91 and the delivery destination hint information 15 that describes the purpose of delivery for each business and its implementation status.

The tweet message delivery means 94 delivers the generated message body 951 to the delivery destination of the delivery destination list 952 at the delivery timing 953.

FIG. 24 shows the detailed configuration of the tweet message delivery destination determination means 93. FIG. 28 shows the operation flow of the tweet message delivery destination determination means 93.

In the tweet message delivery destination determination means 93, first, the situation estimation result 91 and the delivery destination hint information 15 are read (step S601).

As shown in FIG. 25, the situation estimation result 91 is composed of a plurality of accuracy-based situation estimation result elements 911 (eg, test start (0.7), meal start (0.3), condition change response (0.1)). Multiple status estimation result elements with certainty occur when the situation cannot be uniquely identified by the estimation means and uncertainty remains (for example, when the accuracy of the estimated work and execution status is below the threshold value), and a constant value. The work with the above accuracy and the predetermined implementation status are selected. If there is no uncertainty, it will be one. The accuracy may be calculated by using a method such as a Bayesian network or a hidden Markov model in addition to the multi-value logistic regression model and the multi-value logistic regression model as in the first embodiment. Regarding the implementation status, various predetermined implementation statuses such as start, end, and implementation are defined, and the accuracy for each business and each predetermined implementation status is calculated.

The delivery destination hint information 15 is intentionally given by the user, and is additional information for more appropriately selecting the delivery destination. For example, from the mobile terminal 11, the user gives it by a touch panel menu or a voice command. The additional information describes the purpose of distribution such as "report", "emergency", "sharing", "contact", and "record". If it is "report", the boss is selected as the delivery destination, and if it is "record", oneself is selected as the delivery destination. The delivery destination hint information 15 is not an essential requirement.

Next, as shown in FIG. 26, processing is performed using the tweet delivery destination attribute selection rule 934 including a plurality of individual rules 9341. Each individual rule includes a logical expression (IF part) and an attribute (THEN part) composed of each of each situation estimation result element 911 and delivery destination hint information 15. Attributes are used to determine the delivery destination. From these individual rules, all the individual rules for which the logical expression is satisfied are extracted (step S602).

Here, in the IF part of the rule, accuracy is considered in the extraction of attributes. For example, in the case of "start of meal (0.3), response to condition change (0.1)", even if the accuracy of response to condition change is low, if the degree of urgency is high, the delivery destination is extracted on the safety side. An example of the rule is shown below. In this example, the delivery destination hint information is not used.

Example: IF meal start (accuracy&gt; 0.7) THEN Attribute: Nurse Center IF condition change response (accuracy&gt; 0.1) THEN attribute: emergency contact network

Furthermore, the set of attributes described in the THEN part of each extracted individual rule 9341 is extracted (step S603). Here, if there are a plurality of situation estimation result elements with certainty due to uncertainty, the attributes are extracted by OR (logical sum) for both possibilities. This is a strategy to send everything to its destination if there is a possibility of delivery.

Finally, using the delivery destination attribute table 935 as shown in FIG. 27, a list of delivery destinations having each extracted attribute is generated (step S604). In addition, the delivery timing 953 (see FIG. 23) is set for each attribute, and the delivery is performed at that timing.

<img file="JP2012226449A_D0005.tif" />

(Effect of this embodiment) Traditionally, mobile phones and PHS have been used for communication at hospitals, long-term care facilities, stores, maintenance sites, and construction sites. However, in behavioral services, it is troublesome and inconvenient to enter a phone number or e-mail address, and in the case of a call, the work of the recipient is also interrupted, so detailed real-time communication is not possible in reality. there were. On the other hand, in recent years, broadcast-type wireless communication devices called income have become widespread in hospitals, long-term care facilities, and stores. In the income, the channel can be selected, but basically it is delivered to all members belonging to one channel at the same time, and the delivery destination could not be specified depending on the situation and contents. Although it was possible to handle a small number of people, it was difficult when the number of members was large. According to this embodiment, the tweet message can be delivered to the right person at the right time by estimating the situation and hints, and innovates the communication of behavioral services in hospitals, nursing care facilities, stores, maintenance sites, construction sites, etc. It is possible to improve the situation. In particular, appropriate message delivery is possible even when there is uncertainty in the situation estimation.

The business collaboration support system according to the embodiment of the present invention described above can also be realized by using, for example, a general-purpose computer device as basic hardware. That is, each processing unit in the business cooperation support system can be realized by causing the processor mounted on the above-mentioned computer device to execute the program. At this time, the business cooperation support system may be realized by installing the above program in a computer device in advance, storing it in a storage medium such as a CD-ROM, or distributing the above program via a network. Then, this program may be realized by appropriately installing it on a computer device. In addition, as the holding unit and database in the system, the memory, hard disk or storage medium such as CD-R, CD-RW, DVD-RAM, DVD-R, etc. built in or external to the above computer device is appropriately used. Can be realized.

The present invention is not limited to the above-described embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, some components may be removed from all the components shown in the embodiments. In addition, components across different embodiments may be combined as appropriate.