WO2023195117A1 - Group generating device, group generating method, and non-transitory computer-readable medium - Google Patents

Abstract

This group generating device (2000) acquires personal information (10). The personal information (10) represents information pertaining to a plurality of candidate persons. The group generating device (2000) determines the respective allocation of the candidate persons into one or more groups on the basis of a mental state indicator pertaining to the mental state of each candidate person and a generation rule that is a rule for group composition based on the mental state indicators.

Description

Group generation device, group generation method, and non-transitory computer-readable medium

The present disclosure relates to a group generation device, a group generation method, and a non-transitory computer-readable medium.

Technology has been developed to automatically determine the assignment of people to groups. For example, Patent Document 1 discloses a system that automatically determines project group members to meet required specifications.

Japanese Patent Application Publication No. 2007-042059

“Recommended questionnaire for happiness level (SWLS, 4 factors of happiness, etc.)”, [online], Keio University Graduate School of System Design and Management, Human System Design Laboratory, [Retrieved February 16, 2022], Internet < URL: http://lab.sdm.keio.ac.jp/maenolab/questionnaire.html＞

The required specifications disclosed in Patent Document 1 are programming ability, past experience, and time available for the project. Therefore, Patent Document 1 does not disclose a method of determining the assignment of people to groups using indicators other than these. The present disclosure has been made in view of such problems, and one of its purposes is to provide a new technique for assigning people to groups.

The group generation device of the present disclosure includes an acquisition unit that acquires person information regarding a plurality of candidate persons, a psychological state index value regarding the psychological state of each of the candidate persons, and a group composition rule based on the psychological state index value. and determining means for determining assignment of the candidate person to each of one or more groups.

The group generation method of the present disclosure is executed by a computer. The method includes a step of acquiring person information regarding a plurality of candidate persons, a psychological state index value regarding the psychological state of each of the candidate persons, and a group composition rule based on the psychological state index value. and a determining step of determining assignment of the candidate person to each of the above groups.

The non-transitory computer-readable medium of the present disclosure stores a program that causes a computer to execute the information providing method of the present disclosure.

According to the present disclosure, a new technique for assigning people to groups is provided.

FIG. 2 is a diagram illustrating an overview of the operation of the group generation device according to the embodiment. FIG. 2 is a block diagram illustrating a functional configuration of a group generation device. FIG. 2 is a block diagram illustrating the hardware configuration of a computer that implements a group generation device. 3 is a flowchart illustrating the flow of processing executed by the group generation device. 12 is a flowchart showing a procedure for assigning candidate persons according to example 5 of generation rules. FIG. 3 is a diagram illustrating an operation screen provided by the group generation device.

Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding elements are denoted by the same reference numerals, and for clarity of explanation, redundant explanation will be omitted as necessary. Further, unless otherwise specified, predetermined values such as predetermined values and threshold values are stored in advance in a storage device or the like that can be accessed by a device that uses the values. Further, unless otherwise specified, the storage unit is configured by one or more arbitrary number of storage devices.

<Summary>
FIG. 1 is a diagram illustrating an overview of a group generation device 2000 according to an embodiment. Here, FIG. 1 is a diagram for easy understanding of the outline of the group generation device 2000, and the operation of the group generation device 2000 is not limited to that shown in FIG. 1.

The group generation device 2000 determines persons to be assigned to each of a plurality of groups. Hereinafter, a candidate for a person to be assigned to a group will be referred to as a "candidate person."

For example, the group generation device 2000 can be used to divide multiple people belonging to one organization into multiple groups. For example, it is conceivable to use the group generation device 2000 for personnel allocation such as "assigning each of 200 employees to one of five departments." In this case, each employee belonging to this company becomes a candidate person.

In addition, for example, the group generation device 2000 can be used to extract a predetermined number of people from a plurality of people belonging to one organization and create one or more new groups. For example, it is conceivable to use the group generation device 2000 for team formation such as ``selecting 10 employees from among 50 candidate employees to form a new project team.'' In this case, each of the 50 candidates is a candidate. The group generation device 2000 generates a project team by selecting 10 appropriate candidates from 50 candidates.

Note that when a group is generated by selecting some people from a plurality of people in this way, the number of groups that are generated is not limited to one. For example, suppose that two teams are to be formed: project team A consisting of 10 people and project team B consisting of 15 people. It is also assumed that the employees assigned to these project teams are selected from the 50 employees listed as candidates. In this case, the group generation device 2000 appropriately selects 10 employees to be assigned to project team A and 15 employees to be assigned to project team B from among the 50 candidates. Generate a team.

Furthermore, the candidate persons may include persons who cannot be assigned to any group. For example, in the above-mentioned example of ``forming a project team'', 40 out of 50 candidates will not be assigned to a group.

In order to determine the assignment of candidate persons to groups, the group generation device 2000 obtains person information 10. Person information 10 indicates information regarding each candidate person. For example, the person information 10 shows identification information (employee number, name, etc.) of each candidate person.

The group generation device 2000 uses the person information 10 to determine candidate persons to be assigned to each of one or more groups. Here, the psychological state index value of each candidate person is used to assign the candidate person to the group. The psychological state index value is an index value regarding a person's psychological state. For example, the psychological state index value represents the degree of good psychological state of a person.

Furthermore, generation rules are used to assign candidate persons to each group. The generation rule represents a criterion for group composition based on the psychological state index value. Group generation device 2000 determines candidate persons to be assigned to each group based on the psychological state index value and generation rule of each candidate person.

<Example of effects>
According to the group generation device 2000 of the present embodiment, the allocation of candidate persons to groups is determined using the psychological state index value of each candidate person and the generation rule representing the criteria for group composition based on the psychological state index value. Ru. Therefore, it is possible to allocate people to groups, taking into account the psychological state of each person.

Hereinafter, the group generation device 2000 of this embodiment will be explained in more detail.

<Example of functional configuration>
FIG. 2 is a block diagram illustrating the functional configuration of the group generation device 2000 according to the embodiment. The group generation device 2000 includes an acquisition section 2020 and a determination section 2040. The acquisition unit 2020 acquires person information 10. The determining unit 2040 uses the psychological state index value of each candidate person shown in the person information 10 and the generation rule to determine the assignment of the candidate person to one or more groups.

<Example of hardware configuration>
Each functional component of the group generation device 2000 may be realized by hardware that implements each functional component (e.g., a hardwired electronic circuit), or by a combination of hardware and software (e.g., (e.g., a combination of an electronic circuit and a program that controls it). A case in which each functional component of the group generation device 2000 is realized by a combination of hardware and software will be further described below.

FIG. 3 is a block diagram illustrating the hardware configuration of the computer 1000 that implements the group generation device 2000. Computer 1000 is any computer. For example, the computer 1000 is a stationary computer such as a PC (Personal Computer) or a server machine. In addition, for example, the computer 1000 is a portable computer such as a smartphone or a tablet terminal. Computer 1000 may be a dedicated computer designed to implement group generation device 2000, or may be a general-purpose computer.

For example, each function of the group generation device 2000 is realized on the computer 1000 by installing a predetermined application on the computer 1000. The above-mentioned application is composed of programs for realizing each functional component of the group generation device 2000. Note that the method for acquiring the above program is arbitrary. For example, the program can be obtained from a storage medium (DVD disc, USB memory, etc.) in which the program is stored. In addition, for example, the program can be obtained by downloading the program from a server device that manages a storage device in which the program is stored.

The computer 1000 has a bus 1020, a processor 1040, a memory 1060, a storage device 1080, an input/output interface 1100, and a network interface 1120. The bus 1020 is a data transmission path through which the processor 1040, memory 1060, storage device 1080, input/output interface 1100, and network interface 1120 exchange data with each other. However, the method for connecting the processors 1040 and the like to each other is not limited to bus connection.

The processor 1040 is a variety of processors such as a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an FPGA (Field-Programmable Gate Array). The memory 1060 is a main storage device implemented using RAM (Random Access Memory) or the like. The storage device 1080 is an auxiliary storage device implemented using a hard disk, an SSD (Solid State Drive), a memory card, a ROM (Read Only Memory), or the like.

The input/output interface 1100 is an interface for connecting the computer 1000 and an input/output device. For example, an input device such as a keyboard, and an output device such as a display device are connected to the input/output interface 1100.

The network interface 1120 is an interface for connecting the computer 1000 to a network. This network may be a LAN (Local Area Network) or a WAN (Wide Area Network).

The storage device 1080 stores programs that implement each functional component of the group generation device 2000 (programs that implement the aforementioned applications). The processor 1040 implements each functional component of the group generation device 2000 by reading this program into the memory 1060 and executing it.

The group generation device 2000 may be realized by one computer 1000 or by multiple computers 1000. In the latter case, the configurations of each computer 1000 do not need to be the same and can be different.

<Processing flow>
FIG. 4 is a flowchart illustrating the flow of processing executed by the group generation device 2000 of the embodiment. The acquisition unit 2020 acquires person information 10 (S102). The determining unit 2040 determines candidate persons to be assigned to each of one or more groups based on the psychological state index value of each candidate person and the generation rule (S104).

<Acquisition of person information 10: S102>
The acquisition unit 2020 acquires person information 10 (S102). Person information 10 indicates information regarding each candidate person. As described above, the person information 10 indicates identification information of each candidate person. In addition, for example, the person information 10 may further indicate information regarding the attributes of each candidate person. The attributes of the candidate person are used, for example, when predicting a psychological state index value based on the attributes, as described later. Details regarding the attributes of the candidate person will be described later.

There are various specific methods by which the acquisition unit 2020 acquires the personal information 10. For example, the personal information 10 is stored in advance in a storage section that is accessible from the group generation device 2000. In this case, the acquisition unit 2020 acquires the personal information 10 by reading the personal information 10 from this storage unit. In addition, for example, the acquisition unit 2020 may acquire the personal information 10 by receiving the personal information 10 transmitted from another device.

The method of generating the personal information 10 is arbitrary. For example, assume that all people who belong to a specific organization such as a company are treated as candidate people. In this case, the person information 10 can be generated by acquiring information about all the people belonging to the organization using any method. Information about people who belong to a specific organization can be obtained from a system used to manage people who belong to an organization, such as an employee database.

The person information 10 may be generated to indicate only people who meet specific conditions as candidate people. In this case, the personal information 10 can be generated by, for example, searching for a person who meets specific conditions from the aforementioned employee database or the like. Various conditions can be used as conditions, such as age, gender, current or past affiliation, position, qualifications, career history, or work experience.

In addition, for example, the person information 10 may be generated to indicate a person selected by the user of the group generation device 2000 as a candidate person. In this case, for example, the user selects an arbitrary person using the aforementioned employee database or the like. Information indicating each selected person is then used as person information 10.

In addition, for example, it is assumed that the assignment of candidate persons to each group is realized by exchanging members among a plurality of groups. In this case, the candidate persons are existing members of each group. Therefore, information indicating existing members of each group may be used as the person information 10.

For example, assume that members are replaced between department A and department B. That is, assume that some of the existing members of department A are transferred to department B, and some of the existing members of department B are transferred to department A. In this case, the candidate persons are each existing member of department A and each existing member of department B. Therefore, as the person information 10, information indicating existing members of department A and department B can be used.

Note that some, rather than all, existing members may be treated as candidates. For example, suppose that a group includes members who can be moved to other groups and members who cannot be moved to other groups. In this case, among the existing members of the group, members who can be moved to another group are treated as candidate persons. For example, the person information 10 indicates only those members who can be moved to another group among the existing members of the group.

<About psychological state index values>
The psychological state index value is an index value regarding a person's psychological state. Here, the psychological state index value may be expressed as a single index value regarding a person's psychological state, or as a combination of multiple index values regarding a person's psychological state (for example, a vector listing index values). may be done. In the latter case, each index value included in the psychological state index value represents a person's psychological state from a different perspective.

For example, the psychological state index value represents the goodness of a person's psychological state. For example, happiness can be used as a value representing the good psychological state.

A person's happiness level is determined using, for example, the person's answers to questions regarding the four factors of happiness disclosed in Non-Patent Document 1. Non-Patent Document 1 discloses 16 questions for measuring a person's level of happiness. When using the 16 questions related to the four factors of happiness disclosed in Non-Patent Document 1, for example, the psychological state index value is a combination of the answer values for each of these 16 questions (i.e., the combination of the 16 index values). combination). However, instead of using the answers to all 16 questions, only the answers to some of the questions may be used.

In addition, for example, the psychological state index value may be expressed as a combination of scores for each of the four factors of happiness, calculated using answers to questions regarding the four factors of happiness. For example, the 16 questions disclosed in Non-Patent Document 1 include four questions for each of the first to fourth factors of happiness. For example, as psychological state index values, statistical values of answers to four questions regarding the first factor of happiness, statistical values of answers to four questions regarding the second factor of happiness, and answers to four questions regarding the third factor of happiness are used. A combination of four scores can be used: the statistical value of , and the statistical value of the answers to the four questions regarding the fourth factor of happiness. Here, various types of statistical values such as a total sum and an average value can be used as the statistical value.

<About the group>
Various groups can be handled as the groups to which candidate persons are assigned. For example, the meaning of each group is not particularly limited, and only the total number of groups may be specified. In this case, the assignment of candidate persons to each group corresponds to the process of "dividing the plurality of candidate persons shown in the personal information 10 into a specified number of groups."

In addition, for example, each group to which the candidate person is to be assigned may be specified. For example, assume that assignment of candidate persons to two departments, Department A and Department B, is determined. In this case, groups called department A and department B are specified.

For example, the acquisition unit 2020 acquires group information that is information regarding a group to which a candidate person is assigned. For example, group information indicates the total number of groups. In addition, for example, the group information indicates group identification information. In addition, for example, if a group has existing members, the group information indicates identification information of the existing members of the group.

In addition, for example, the group information may indicate constraint conditions regarding the group. For example, the constraint condition is a condition regarding the minimum number of people to be included in the group, the maximum number of people, or both. In addition, for example, the constraint condition is a condition regarding the attributes of a person to be included in a group.

For example, assume that the attribute is a level that is set according to a person's ability or experience. As the level, for example, a level that allows the user to work as a manager (hereinafter referred to as manager level) or a level that allows the user to work as a group leader (hereinafter referred to as group leader level) can be adopted. When levels are treated as attributes in this way, for example, the constraint conditions for a group are conditions regarding the minimum and maximum number of people of each level to be assigned to the group. For example, it is conceivable to set a constraint such as "one or more people are at the manager level and three or more people are at the group leader level."

Here, when dealing with multiple groups, the constraint conditions may be common to the multiple groups or may be different for each group. Moreover, both a constraint condition common to a plurality of groups and a constraint condition different for each group may be set.

<Determination of assignment of candidate person: S104>
The determining unit 2040 determines the assignment of candidate persons to each group based on the psychological state index value of each candidate person and the generation rule. The generation rule is a standard for determining candidate persons to be assigned to each group based on the psychological state index value. For example, if the process of determining candidate persons to be assigned to each group is considered as the process of solving a combinatorial optimization problem, the generation rule can be expressed as an objective function to be maximized or minimized in the combinatorial optimization problem. . Therefore, candidate persons to be assigned to each group are determined by combinatorial optimization in which candidate persons are assigned to each group so as to maximize or minimize the objective function.

When the generation rule is expressed by maximizing or minimizing an objective function, for example, the determining unit 2040 uses various algorithms for solving a combinatorial optimization problem to maximize or minimize the objective function. Next, the assignment of candidate persons to each group is determined. However, the allocation of candidate persons to each group that maximizes or minimizes the objective function, which is determined by the determining unit 2040, does not have to be an exact solution. For example, the determining unit 2040 may determine the assignment of candidate persons to each group so as to maximize or minimize the objective function using various algorithms for finding approximate optimal solutions to the combinatorial optimization problem.

Furthermore, as described later, the generation rule does not necessarily need to be expressed as an objective function.

Hereinafter, some specific examples of generation rules will be illustrated.

　ここで、G はグループの集合を表し、i はグループの識別子を表し、N はグループの総数を表す。集合 G は、G={G[i]|1<=i<=N} と表すことができる。なお、グループの総数が１である場合、N=1 である。 <<Generation rule example 1>>
For example, the generation rule is a rule that "both candidate persons with large psychological state index values and persons with small psychological state index values are assigned to a group." For example, this generation rule can be expressed as a rule that maximizes the objective function U1(G) below.

Here, G represents the set of groups, i represents the group identifier, and N represents the total number of groups. The set G can be expressed as G={G[i]|1<=i<=N}. Note that when the total number of groups is 1, N=1.

G[i] represents a set of identifiers of people included in the i-th group. Note that the group identifier used in equation (1) and each mathematical equation described below is a natural number uniquely assigned to each group in order from 1. Similarly, in equation (1) and each mathematical expression described later, the identifier of the person included in G[i] is a natural number that is uniquely assigned to each person included in G[i] in order from 1.

M represents the number of dimensions of the psychological state index value when the psychological state index value is represented by a multidimensional vector. For example, when a vector listing answers to each of the 16 questions regarding the four factors of happiness disclosed in Non-Patent Document 1 is used as a psychological state index value, M=16. Note that when the psychological state index value is represented by one value, M=1. j is the index of an element in the vector representing the psychological state index value.

S[i][j] represents a set obtained by extracting only the value of the j-th element from the psychological state index values of each person included in group i. k is an identifier of a person included in group i. v[k] represents the psychological state index value of person k. v[k][j] represents the value of the j-th element of vector v[k] representing psychological state index values.

In D1(i) of equation (1), the difference between the maximum and minimum psychological state index values in group i is calculated for each element j. Then, the sum of the differences between the maximum and minimum values calculated for each element j is calculated as D1(i). Then, the sum of D1(i) calculated for each group i is calculated as U1(G).

The determining unit 2040 determines the set G that maximizes the objective function U1(G). To this end, for example, the determining unit 2040 calculates the value of the objective function U1(G) while varying the assignment of candidate persons to each group i (in other words, the identification information of the persons included in G[i]), By comparing the calculated values, determine the set G that maximizes the objective function U1(G). Here, various existing algorithms can be used as a specific algorithm for determining the allocation of elements (in this case, candidate persons) to a plurality of groups so that the objective function is maximized. Note that if there are existing members in group i and those members cannot be moved to another group, the identifiers of those existing members are always included in G[i].

By maximizing the objective function defined by formula (1), it is possible to increase the difference between the maximum value and the minimum value for each element of the psychological state index value in each group. By doing this, each team can be generated so that members in a good psychological state can support members in a poor psychological state.

Note that part of the objective function U1(G) can be changed without departing from its purpose. For example, U1(G) may be determined by the average of D(i) instead of the sum of D(i). Further, D(i) may be determined by the average of the differences between the maximum and minimum values calculated for each element, instead of the sum of the differences between the maximum and minimum values calculated for each element.

　ここで、A[j] は、任意のグループに含まれる各人物の心理状態指標値から、第 j 要素の値のみを取り出して得られる集合を表す。その他の記号については、式（１）と同様である。 <<Generation rule example 2>>
For example, the generation rule is a rule that says, "The difference between the psychological state index value for a group and the psychological state index value for the whole group should be made small." For example, this generation rule can be expressed as a rule to minimize the following objective function U2(G).

Here, A[j] represents a set obtained by extracting only the value of the j-th element from the psychological state index values of each person included in an arbitrary group. Other symbols are the same as in formula (1).

In D2(i) of Equation (2), the absolute value of the difference between the average value of the psychological state index values in all groups and the average value of the psychological state index values in group i is calculated for each element j. Then, the sum of the absolute values of the differences calculated for each element j is calculated as D2(i). Then, the sum of D2(i) calculated for each group i is calculated as U2(G).

The determining unit 2040 determines the set G that minimizes the objective function U2(G). Here, various existing algorithms can be used as a specific algorithm for determining the assignment of elements (in this case, candidate persons) to a plurality of groups so that the objective function is minimized. Note that if there are existing members in group i and those members cannot be moved to another group, the identifiers of those existing members are always included in G[i].

By minimizing such an objective function, each group is generated such that the difference between the average psychological state index value within the group and the average psychological state index value among all groups becomes small. Therefore, candidate persons are assigned to each group so that the psychological state index values are balanced among the groups.

Note that part of the objective function U2(G) can be changed without departing from its purpose. For example, in the objective function U2(G), an average or the like may be used instead of the sum. Furthermore, in D2(i) , a square or the like may be used instead of the absolute value.

<<Generation rule example 3>>
For example, the generation rule is a rule that "make the difference between the psychological state index value in a group and the psychological state index value for the whole group large." For example, such a generation rule can be expressed as a rule that maximizes the objective function U3(G) below.

　ここで、D3(i) と、式（２）の D2(i) から絶対値記号を外したものである。

Here, D3(i) is obtained by removing the absolute value symbol from D2(i) in equation (2).

The determining unit 2040 determines the set G that maximizes the objective function U3(G). Here, as described above, various existing algorithms can be used as specific algorithms for determining the allocation of elements to a plurality of groups so that the objective function is minimized. Note that if there are existing members in group i and those members cannot be moved to another group, the identifiers of those existing members are always included in G[i].

According to equation (3), a group is generated that has the characteristic that "the average of the psychological state index values within the group is larger than the average of the psychological state index values of the entire group." By doing this, it is possible to generate a strong group with a high degree of good psychological state.

Note that part of the objective function U3(G) can be changed without departing from its purpose. For example, in the objective function U3(G), an average or the like may be used instead of the sum.

　ここで、var(X) は、集合 X の要素の分散を表す。var(Y) についても同様である。集合 X の要素 x[k]は、グループ i に含まれる人物 k について、心理状態指標値を表すベクトル v[k] の全ての要素の平均値を表す。集合 Y の要素 y[k]は、任意のグループに含まれる人物 k について、心理状態指標値を表すベクトル v[k] の全ての要素の平均値を表す。 <<Generation rule example 4>>
For example, the generated rule is a rule that says, "The variation in psychological state index values within a group should be made to be close to the variation in psychological state index values for the whole group." This generation rule can be expressed, for example, as a rule to minimize the objective function U4(G) below.

Here, var(X) represents the variance of the elements of set X. The same goes for var(Y). Element x[k] of set X represents the average value of all elements of vector v[k] representing psychological state index values for person k included in group i. The element y[k] of the set Y represents the average value of all elements of the vector v[k] representing the psychological state index value for the person k included in any group.

The determining unit 2040 determines the set G that minimizes the objective function U4(G). As described above, various existing algorithms can be used as specific algorithms for determining the allocation of elements to multiple groups so that the objective function is minimized. Additionally, if group i has existing members and those members cannot be moved to another group, the identifiers of those existing members are always included in G[i].

According to the generation rule expressed by equation (4), the variation in psychological state index values can be made to be the same in each group. Therefore, it is possible to prevent the generation of a group in which the dispersion in psychological states among members is extremely large.

Note that part of the objective function U4(G) can be changed without departing from its purpose. For example, in the objective function U4(G), an average or the like may be used instead of the sum. Furthermore, in D4(i), a square or the like may be used instead of the absolute value.

<<Generation rule example 5>>
In addition, for example, the generation rule is a rule that "persons with large psychological state index values and persons with small psychological state index values are equally assigned to each group." The assignment of candidate persons according to this generation rule can be expressed, for example, by the procedure shown in FIG. FIG. 5 is a flowchart showing the procedure for assigning candidate persons according to Example 5 of the generation rule.

The determining unit 2040 generates a permutation P in which candidate persons are sorted in descending or ascending order of psychological state index value (S202). The determining unit 2040 initializes the identifier i of the group to be processed to 1 (S204).

S206 to S214 are loop processing L1 executed for each group. In S206, the determining unit 2040 determines whether the identifier of the group to be processed is less than or equal to the total number of groups N (that is, whether i<=N). If i<=N, the loop process L1 ends, and the process in FIG. 5 proceeds to S2204. On the other hand, if i<=N, the process in FIG. 5 proceeds to S208.

In S208, the determining unit 2040 extracts the first person a and the last person b from the permutation P, and assigns them to group i. Here, a and b may be the same value or different values.

The determining unit 2040 determines whether there are any candidate persons remaining in permutation P (S210). If no candidate person remains in the permutation P (S210: NO), the process of FIG. 5 ends.

On the other hand, if a candidate person remains in the permutation P (S210: YES), the determining unit 2040 adds 1 to i (S212). Next, since S214 is the end of the loop process L1, the process in FIG. 5 proceeds to S206.

In this way, according to the series of processes represented by the flowchart in FIG. Repeated. Therefore, a person with a large psychological state index value and a person with a small psychological state index value are assigned to each group in order. By assigning candidate persons in this manner, it is possible to balance the psychological states of each group.

<Method to identify psychological state index value>
In order to determine the assignment of candidate persons to groups using the method described above, it is necessary to specify the psychological state index value v[k] for each candidate person k. For example, the psychological state index value is predetermined for each candidate person. In this case, the determining unit 2040 uses a predetermined psychological state index value for each candidate person. For example, in this case, the person information 10 indicates, for each candidate person, the identification information of the candidate person and the psychological state index value in association with each other.

The method of predetermining the psychological state index value is arbitrary. For example, assume that the psychological state index value is determined using the answers to the questions regarding the four factors of happiness described above. In this case, for example, each candidate (e.g., each employee) is sent a questionnaire asking questions regarding the four factors of happiness, and responses to the questionnaire are collected from each candidate to determine the psychological state of each candidate. Index values can be calculated.

Here, the psychological state index value of each candidate person may be calculated from the results of a single questionnaire (for example, the results of the most recent questionnaire), or may be calculated from the results of multiple questionnaires. In the latter case, for example, the above-mentioned questionnaire is periodically sent to each person. Thereby, the psychological state index value can be calculated for each person from each of the plurality of questionnaires. Therefore, for example, statistical values of a plurality of psychological state index values obtained using a plurality of questionnaires are used as the psychological state index value of each person. For example, if a questionnaire is conducted once a week, a method may be considered in which the psychological state index value of each person is calculated using the results of the last four questionnaires.

The psychological state of each candidate may change depending on the group to which they are assigned. For example, a person's psychological state is likely to be significantly different depending on whether he or she is assigned to a group with a boss with whom he or she is bad, or whether he or she is assigned to a group with a boss he or she gets along well with. Additionally, a person's psychological state is thought to be significantly different between a case where the person is assigned to a department related to the job they desire and a case where the person is assigned to a department related to the job they do not desire. .

Therefore, for example, when trying out various patterns for assigning candidate people to each group, the determining unit 2040 calculates the psychological state index value of each candidate person in each trial based on the composition of the group in that trial. Good too. In this case, for example, an estimation model for estimating the psychological state index value of each candidate person is determined in advance. The estimation model can be realized by any machine learning model such as a neural network.

An estimation model that estimates a psychological state index value of a certain person is configured to output a psychological state index value of a certain person in response to input of information representing the environment in which the person is placed, for example. . A person's environment is represented by, for example, an attribute of the person, an attribute of a group to which the person belongs, or an attribute of each person belonging to the same group as the person. Here, as the attributes of a person that represent the person's environment, attributes that change depending on the group to which the person belongs, such as job or occupation, are used. As the attributes of the group, information representing the characteristics of the group can be used, such as the location of the group's office or the like, or the frequency of overtime work or work on holidays in the group. As for the attributes of each person belonging to the same group, various attributes such as personality, age, gender, current or past department, position, qualifications, career, or work experience can be used.

In order to generate an estimation model for each person, for example, for each person, the history of past psychological state index values is matched with information representing the environment in which that person was placed when the psychological state index value was calculated. and record it. The estimation model is trained using a plurality of training data consisting of pairs of "input data = information representing the environment, correct data = psychological state index value".

<Consideration of constraints>
If the group information indicates a constraint regarding the group, the determining unit 2040 considers the constraint indicated in the group information in addition to the generation rule to determine assignment of the candidate person to the group. Here, various existing algorithms are used as specific algorithms for determining elements (candidate persons in this disclosure) to be assigned to each group so as to maximize or minimize the objective function while taking into account constraints. can be used.

Furthermore, as shown in example 5 of generation rules, even when an objective function is not used, candidate persons can be assigned in consideration of constraint conditions using various methods. For example, assume that the constraint condition is a condition regarding a person's attributes. In this case, when extracting candidate persons from the permutation P in S208, the determining unit 2040 preferentially extracts persons satisfying the constraint conditions. For example, assume that a constraint condition is set such that each group includes at least one manager level person. In this case, for example, before executing the loop process L1, the determining unit 2040 determines, for each group, that ``the manager-level candidate person who is closest to the beginning of the permutation P is extracted from the permutation P and added to the group.'' Execute processing in order.

<Result output>
Group generation device 2000 outputs output information representing processing results. For example, the output information indicates, for each of one or more groups, the identification information of the group and the identification information of each candidate person assigned to the group.

In addition, for example, if the psychological state index value of each candidate person is estimated according to the group to which the candidate person is assigned, the output information may include the estimated psychological state index value for each candidate person assigned to the group. An index value may also be indicated. Furthermore, the output information may indicate, for each group, statistical information regarding the psychological state index value in that group. The statistical information regarding the psychological state index value indicates, for example, the maximum value, minimum value, average, or variance of the psychological state index value in the group.

The output mode of the output information is arbitrary. For example, the output information is stored in any storage unit. For example, the output information may also be displayed on any display device. For example, the output information may also be transmitted to any device.

<Example of operation screen provided by group generation device 2000>
Here, an example of an operation screen provided by the group generation device 2000 will be illustrated. Here, the functional component that outputs the operation screen is referred to as a screen output section. FIG. 6 is a diagram illustrating an operation screen provided by the group generation device 2000. The operation screen 100 in FIG. 6 includes a display area 110, a display area 120, a display area 130, and a display area 140.

In this example, the group generation device 2000 is used to allocate personnel to each group under the division named Division A. For example, a user of the group generation device 2000 selects a business whose personnel allocation is desired to be changed from among a plurality of business divisions owned by the company. In response to this selection, the group generation device 2000 acquires person information 10 indicating all the people belonging to the selected business division as candidate persons. The group generation device 2000 also acquires group information indicating information on each group under the selected business division. Then, the group generation device 2000 assigns each person belonging to the selected business division to one of the groups under the selected business division.

In this example, the target is Division A. Therefore, the group generation device 2000 assigns each person belonging to the A division to any one of the G1 group, G2 group, and G3 group under the A division. As a result of the assignment, the group generation device 2000 outputs the operation screen 100 shown in FIG. 6.

The display area 110 shows the average value of the psychological state index values of all the people calculated for the entire A division (that is, all groups under the A division). Here, a value estimated using the above-mentioned estimation model is used as the psychological state index value of each person. In this example, the psychological state index value is expressed as the sum of scores calculated for each of the four happiness factors disclosed in Non-Patent Document 1. In the display area 110 and the display area 120, graphs representing the magnitude of the psychological state index value are obtained by accumulating the scores of each factor.

In the display area 110, the "original plan" indicates the average value of the psychological state index values of all the people in the case where personnel allocation is performed as determined by the group generation device 2000. As will be described later, the "comparison plan" indicates the average value of the psychological state index values of all the people in the case where the assignment of the people is changed by the user's input operation. The input operation by the user will be described later. "Current" indicates the average value of all psychological state index values for the current actual group composition. "Last time" indicates the average value of all psychological state index values for the past actual group composition.

The display area 120 shows the configuration of the A division. By selecting two groups displayed in the display area, the user can exchange members between the two selected groups. The above-mentioned "comparative plan" shows information about a case where the personnel allocation is changed by this replacement operation.

The display area 130 shows, for each of the two selected groups, the average value of the psychological state index values of all the people belonging to that group.

After viewing the configuration of the group automatically generated by the group generation device 2000, the user can attempt to change the personnel allocation by performing the above-described replacement operation. Then, by comparing the psychological state index value before the trial (original draft) and the psychological state index value after the trial (comparison draft), the staffing arrangement automatically generated by the group generation device 2000 can be modified. It is possible to judge whether or not it is appropriate.

Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above embodiments. The configuration and details of the present invention can be modified in various ways that can be understood by those skilled in the art within the scope of the present invention.

Note that in the above example, the program includes a set of instructions (or software code) for causing the computer to perform one or more of the functions described in the embodiments when loaded into the computer. The program may be stored on a non-transitory computer readable medium or a tangible storage medium. By way of example and not limitation, computer readable or tangible storage media may include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD - Including ROM, digital versatile disc (DVD), Blu-ray disc or other optical disc storage, magnetic cassette, magnetic tape, magnetic disc storage or other magnetic storage device. The program may be transmitted on a transitory computer-readable medium or a communication medium. By way of example and not limitation, transitory computer-readable or communication media includes electrical, optical, acoustic, or other forms of propagating signals.

Part or all of the above embodiments may be described as in the following additional notes, but are not limited to the following.
(Additional note 1)
an acquisition means for acquiring personal information regarding a plurality of candidate persons;
determining means for determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; Group generation device with.
(Additional note 2)
The group generation device according to supplementary note 1, wherein the person information indicates the psychological state index value of each candidate person.
(Additional note 3)
An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
The group generation device according to supplementary note 1, wherein the determining means calculates the psychological state index value of each candidate person using the estimation model.
(Additional note 4)
The group generation device according to any one of Supplementary Notes 1 to 3, wherein the rule is a rule that maximizes or minimizes an objective function calculated from the psychological state index value of each person included in each group. .
(Appendix 5)
The person information includes information regarding attributes of each candidate person,
The group generation device according to any one of Supplementary Notes 1 to 3, wherein the determining means determines the assignment of the candidate person to each group so as to satisfy a constraint regarding an attribute of the candidate person to be included in each group. .
(Appendix 6)
For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. have the means,
The screen output means, in response to an operation of exchanging a person included in the first group and a person included in the second group, displays information in the first group after the exchange operation. Supplementary Notes 1 to 3, wherein the statistical values of the psychological state index values of the included persons and the statistical values of the psychological state index values of the persons included in the second group after the replacement operation are displayed on the operation screen. The group generation device according to any one of the items.
(Appendix 7)
an acquisition step of acquiring person information regarding a plurality of candidate persons;
a determining step of determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; A computer-implemented group generation method comprising:
(Appendix 8)
The group generation method according to appendix 7, wherein the person information indicates the psychological state index value of each candidate person.
(Appendix 9)
An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
The group generation method according to appendix 7, wherein in the determining step, the psychological state index value of each candidate person is calculated using the estimation model.
(Appendix 10)
The group generation method according to any one of Supplementary Notes 7 to 9, wherein the rule is a rule that maximizes or minimizes an objective function calculated from the psychological state index value of each person included in each group. .
(Appendix 11)
The person information includes information regarding attributes of each candidate person,
The group generation method according to any one of Supplementary Notes 7 to 9, wherein in the determining step, the assignment of the candidate person to each group is determined so that constraint conditions regarding the attributes of the candidate person to be included in each group are satisfied. .
(Appendix 12)
For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. has a step,
In the screen output step, in response to an operation of exchanging a person included in the first group and a person included in the second group, the information is displayed in the first group after the exchange operation. Supplementary Notes 7 to 9, wherein the statistical values of the psychological state index values of the included persons and the statistical values of the psychological state index values of the persons included in the second group after the replacement operation are displayed on the operation screen. The group generation method according to any one of the items.
(Appendix 13)
an acquisition step of acquiring person information regarding a plurality of candidate persons;
a determining step of determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; A non-transitory computer-readable medium that stores a program that is executed by a computer.
(Appendix 14)
The computer-readable medium according to appendix 13, wherein the person information indicates the psychological state index value of each candidate person.
(Additional note 15)
An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
14. The computer-readable medium according to appendix 13, wherein in the determining step, the psychological state index value of each candidate person is calculated using the estimation model.
(Appendix 16)
The computer-readable medium according to any one of appendices 13 to 15, wherein the rule is a rule for maximizing or minimizing an objective function calculated from the psychological state index value of each person included in each of the groups. .
(Appendix 17)
The person information includes information regarding attributes of each candidate person,
The computer-readable medium according to any one of Supplementary Notes 13 to 15, wherein in the determining step, the assignment of the candidate person to each group is determined so as to satisfy the constraint regarding the attributes of the candidate person to be included in each group. .
(Appendix 18)
For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. has a step,
In the screen output step, in response to an operation of exchanging a person included in the first group and a person included in the second group, the information is displayed in the first group after the exchange operation. Supplementary Notes 13 to 15, wherein the statistical values of the psychological state index values of the included persons and the statistical values of the psychological state index values of the persons included in the second group after the replacement operation are displayed on the operation screen. A computer-readable medium according to any one of the preceding paragraphs.

10 Person information 100 Operation screen 110 Display area 120 Display area 130 Display area 140 Display area 1000 Computer 1000 Each computer 1020 Bus 1040 Processor 1060 Memory 1080 Storage device 1100 Input/output interface 1120 Network interface 2000 Group generation device 2020 Acquisition unit 2040 Determination unit

Claims (18)

an acquisition means for acquiring personal information regarding a plurality of candidate persons;
determining means for determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; Group generation device with. The group generation device according to claim 1, wherein the person information indicates the psychological state index value of each candidate person. An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
The group generation device according to claim 1, wherein the determining means calculates the psychological state index value of each candidate person using the estimation model. Group generation according to any one of claims 1 to 3, wherein the rule is a rule that maximizes or minimizes an objective function calculated from the psychological state index value of each person included in each group. Device. The person information includes information regarding attributes of each candidate person,
The group generation method according to any one of claims 1 to 3, wherein the determining means determines the assignment of the candidate person to each group so as to satisfy a constraint regarding an attribute of the candidate person to be included in each group. Device. For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. have the means,
The screen output means, in response to an operation of exchanging a person included in the first group and a person included in the second group, displays information in the first group after the exchange operation. The statistical value of the psychological state index value of the included person and the statistical value of the psychological state index value of the person included in the second group after the replacement operation are displayed on the operation screen. 3. The group generation device according to any one of 3. an acquisition step of acquiring person information regarding a plurality of candidate persons;
a determining step of determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; A computer-implemented group generation method comprising: The group generation method according to claim 7, wherein the person information indicates the psychological state index value of each candidate person. An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
8. The group generation method according to claim 7, wherein in the determining step, the psychological state index value of each candidate person is calculated using the estimation model. Group generation according to any one of claims 7 to 9, wherein the rule is a rule that maximizes or minimizes an objective function calculated from the psychological state index value of each person included in each group. Method. The person information includes information regarding attributes of each candidate person,
Group generation according to any one of claims 7 to 9, wherein in the determining step, assignment of the candidate person to each group is determined so as to satisfy constraint conditions regarding attributes of the candidate person to be included in each group. Method. For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. has a step,
In the screen output step, in response to an operation of exchanging a person included in the first group and a person included in the second group, the information is displayed in the first group after the exchange operation. From claim 7, the statistical value of the psychological state index value of the included person and the statistical value of the psychological state index value of the person included in the second group after the replacement operation are displayed on the operation screen. 9. The group generation method according to any one of 9. an acquisition step of acquiring person information regarding a plurality of candidate persons;
a determining step of determining the assignment of the candidate person to each of one or more groups based on a psychological state index value regarding the psychological state of each candidate person and a group composition rule based on the psychological state index value; A non-transitory computer-readable medium that stores a program that is executed by a computer. The computer-readable medium according to claim 13, wherein the person information indicates the psychological state index value of each candidate person. An estimation model is defined for estimating the psychological state index value of the candidate person according to the environment of the group to which the candidate person belongs,
14. The computer-readable medium according to claim 13, wherein in the determining step, the estimation model is used to calculate the psychological state index value of each candidate person. The computer-readable method according to any one of claims 13 to 15, wherein the rule is a rule that maximizes or minimizes an objective function calculated from the psychological state index value of each person included in each group. Medium. The person information includes information regarding attributes of each candidate person,
The computer-readable method according to any one of claims 13 to 15, wherein, in the determining step, assignment of the candidate person to each group is determined so as to satisfy constraints regarding attributes of the candidate person to be included in each group. Medium. For each of the first group and the second group, a screen output that outputs an operation screen showing statistical values of the psychological state index values of the people included in the group and information on each person included in the group. has a step,
In the screen output step, in response to an operation of exchanging a person included in the first group and a person included in the second group, the information is displayed in the first group after the exchange operation. From claim 13, wherein the statistical value of the psychological state index value of the included person and the statistical value of the psychological state index value of the person included in the second group after the replacement operation are displayed on the operation screen. 15. The computer readable medium according to any one of 15.

Images