JP2000250989A - Worker allocation system, worker allocating method and storage medium readable by computer in which worker allocation program is recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a worker allocation system to allocate workers based on normal execution probability to normally execute each process to constitute a work. SOLUTION: This system is constituted by providing an input part 1 to input process names of the processes to constitute the work to be executed and information about requested ability to be requested as an ability to execute the processes, a worker information storage part 3 in which names of plural workers to execute the processes and information about an ability of the worker about a process execution ability of each worker are stored, a probability calculating part 8 to calculate the normal execution probability to normally complete each process by each worker based on the inputted information of the requested ability and information about the ability of the worker stored in the worker information storage part 3 and a work optimizing part 9 to allocate the workers to execute the processes based on the calculated normal execution probability.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

TECHNICAL FIELD The present invention relates to a worker allocating system, a worker allocating method, and a work for allocating an optimum worker (or a work group or a work company) from candidates in a work composed of one or a plurality of processes. The program for assigning employees.

[0002]

2. Description of the Related Art In recent years, the industry has been subjected to severe international competition. In this environment, the products or services to be provided compete with each other for essential superiority in novelty, function, quality, price, distribution system, service system, and the like.
As one of the measures for a company to cope with such a competitive environment, not all processes related to the products or services to be provided are performed within the company itself, but resources outside the company (human resources, (Parts, semi-finished products, equipment, etc.) and integrate the optimal parts and services from inside and outside the company, the resulting products or services will be much more competitive than others. Would. The utilization of these external resources has been further promoted, and a part of the processes that were previously closed within the company has been shared between multiple different companies as necessary, and cooperation between the companies has been carried out, as if it were a single company. Is a virtual enterprise (NIIIP (National)
Industrial Information I
nfrastructure Protocols),
1996, http: // www. niiip. org
/ Public-form /).

[0003] Unlike procurement of standard parts, procurement of a process that is difficult to clarify the procurement contents, such as design work, from the outside involves a great risk of procurement. Companies have expanded the scope and scale of their processes by incorporating them into their own companies (Kenichi Imai, Takayuki Itami, Kazuo Koike, "Economics of Internal Organizations", Toyo Keizai Shimpo, Tokyo, 1
992). Problems associated with large-scale enterprises have been pointed out as excessive division of labor and lack of communication between departments (translated by Naoya Yoda, "Made in Americanic").
a ", Soshisha, 1990, MIT Industrial Productivity Investigation Committee), (Hammer, M., Champy, J.,
“Reengineering A Manifest
o forBusiness Revolution
n, "Linda Michaels Literary
(Agency, translated by Ikujiro Nonaka, "Reengineering Revolution", Nihon Keizai Shimbun, Tokyo, 1993), and these problems are also factors in the necessity of a new corporate form.
On the other hand, the difficulty of external procurement is generally analyzed in terms of the next transaction cost in commercial transactions (for example, Toshiyuki Yahagi, "Modern Distribution", Alma Yuhikaku, Tokyo, 1996).
The search cost is a cost for specifying a supplier.
Negotiation costs are costs associated with negotiating contract terms.
The monitoring cost is a cost for monitoring whether the monitoring is performed according to the contract. It should be understood that this concept of monitoring costs also includes the cost of recovering progress if the contract does not progress as planned. Virtual enterprises are generally a form of corporate collaboration in which companies do not have long-lasting business relationships, and the above transaction costs must be reduced to a reasonable level in order for this corporate form to become universal in the industry. Various social conditions need to be established in order to ensure that In particular, prior to the procurement contract, "whether to procure a process or semi-finished product internally or externally (make
), a rational planning function and a monitoring / control function after the contract are required. In other words, it is necessary to consider the risk assessment in advance when the procurement from the contractor does not progress as planned and the measures to be taken when a risk occurs after the execution of the process.

[0004] Regarding risk management, market risk and credit risk are used in the field of finance (Shoichi Ninomiya, Rei Tajima, Hideyuki Mizuta, "The Forefront of IT in Financial Risk Management", Information Processing, Volume 39, Issue 8, 1
998, p. 794-799). The risk in this statement is similar to credit risk in the financial sector, but credit risk is in terms of assessing the value of the financial product by the credit of the company issuing the product, and the risk in this statement is This is a perspective on the risk of not obtaining the planned results when outsourcing, and both perspectives are different.

Next, an overview of the definition of a virtual enterprise will be given. In the United States, the Internet, S
TEP (Standard for the Exch)
angel of Product Model Dat
a), CORBA (Common Object Re)
quest Broker Architecture
e), a national budget project NIIP (National Indus) that aims to realize a virtual enterprise using the four element technologies of workflow management.
trial Information Infrast
The Protocol (NRT96-01) of R.T.R.ture Protocols (NRT96-01) defines a virtual enterprise as follows (NIIIP, 199).
6). Virtual Enterprise is
a temporal consortium or
alliance of companies fo
rmed to share costs and s
kills andexploit fast-cha
ng opportunities. Translated this way, "The virtual enterprise is a temporary consortium or coalition of companies to share costs, skills, seize the fast-changing opportunities, and conduct business activities." What should be done is that, unlike conventional corporate partnerships based on capital series or individual negotiations, unspecified (although under some kind of registration system) firms will collaborate as needed, and the partnership will be dissolved when it is no longer needed This is a form of corporate cooperation. In that sense, it differs from the conventional corporate form, such as a joint venture, which requires a lot of negotiating items individually before the corporate alliance.

[0006]

Problems to be solved by the virtual enterprise will be described below. In order for the virtual enterprise to become universal, it is necessary to solve the following interrelated technical and non-technical issues. First, there is a need for a common enterprise reference model and terminology between companies. When establishing a virtual enterprise, it is desirable to obtain in advance a common understanding of the processes and contents shared among member companies in order to reduce the matters to be considered individually as much as possible, that is, to reduce negotiation costs. for that reason,
Standardizing reference models and terminology is effective. As a representative study of the reference model, ISO TC184
Then, ISO (International Organa)
nizationforStandardizatio
n) In 14258 (CD), PERA (Purd
ue Enterprise Reference A
An enterprise model is being studied based on the (raturecure) method (ISO 1997, http.
p: // www. mel. nest. gov / sc5w
g1 / std-drft. htm). Besides ISO, I
MS (IntelligentManufacturi)
ng System) Project (IMS 199
8, http: // ims. toyo-eng. co.
jp / pub / gm21pub / h8_abst_e
/ Japan Machinery Federation (Nippon Machinery Federation, 1997, Japan Machinery Federation “Survey and research on standardization of virtual enterprises (Part 1)” 1996 report, 1997, Japan Machinery Federation,
1998, Japan Machinery Federation, "Investigation and research on standardization of virtual companies (part 2)", 1997 report, 1
998). Second, it is necessary to establish a corporate information disclosure base. That is, in order to reduce transaction search costs, it is necessary to have an information base for registering virtual enterprise configuration candidate member companies having certain requirements and disclosing the information. Third, it is necessary to improve the integrated data environment. After the establishment of the virtual enterprise, management data, design data,
An integrated data environment (IDE: Integr) with security functions such as prohibiting access from anyone other than the contractor in order to share manufacturing data etc. among member companies.
attached Data Environment) is required. Currently, the U.S. Department of Defense has JCALS (Joint)
Continuous Acquisition a
ndLife-cycle Support) project (JCALS, 1998,
http://150.149.1.11/segme
nted_products / csc / capr3_1
It is desired that the standardization of (/) be applied to data sharing in the cooperative design work process between unspecified companies for private use. In Japan, NCCALS (Integrated Information Technology Research Association for Production, Procurement and Operation Support) and its successor JECALS (Japan Electronic Commerce Promotion Organization (http: //www.jecals.jpde)
c. or. jp /)). Fourth, virtual enterprise establishment and operation technology is required. It is necessary to collaborate with a company that has been registered but has no long-term continuous business relationship as if it were a single company, albeit temporarily, so trust in candidate members (workers and companies). Problem, that is, a method of risk assessment is needed. In addition, a method for monitoring and controlling the operation of the virtual enterprise after the contract is needed. As one of the trends, a TEMPLET project has been examining a framework for expressing how companies are responding to the virtual enterprise era, mainly in Canada (http://www.templet.com). org).
Fifth, it is necessary to establish a legal system. Establish a legal system that allows the virtual enterprise to legally act as a legal entity and perform contracts and other legal acts, and to take on legal responsibilities in accordance with the divisions of the companies that make up the virtual enterprise is necessary.
It is necessary to review the tax system, commercial law, and other legal systems related to intellectual property rights and manufacturer responsibilities so that the new enterprise form of virtual enterprise can be applied.
Sixth, a mediation organization needs to be established. There is a need for an organization that notarizes the contract of the virtual enterprise and, after the work is completed, arbitrates if the work is not performed according to the contract. As described above, there are many issues for the realization and diffusion of the virtual enterprise, but this invention is one of the technical issues of the establishment and operation of the virtual enterprise described above. It aims to solve the problem about.

[0007]

A worker assignment system according to the present invention uses, as input information, a process name of a process constituting a task to be performed and information on a required capability required as a capability of performing the process. An input unit for inputting, an input information storage unit for storing input information input by the input unit, worker names of a plurality of workers who are candidates for performing the process, and a process execution capability of each worker. A worker information storage unit that stores information about the worker ability, based on the information about the required ability stored in the input information storage unit and the information about the worker ability that is stored in the worker information storage unit, Arithmetic processing in which each worker calculates a normal performance probability of successfully completing the above process, and assigns workers performing the process based on the calculated normal performance probability. And an output information storage unit for storing the worker name of the worker assigned by the arithmetic processing unit, and an output unit for outputting the worker name of the worker stored in the output information storage unit. Features.

[0008] The worker allocation system further includes a probability storage unit for storing the normal execution probability. The input unit stores process names of a plurality of processes, information on a plurality of required capabilities corresponding to the processes. The input processing information is input, the arithmetic processing unit calculates a normal execution probability that the worker successfully completes each process, a probability calculation unit that stores the calculated normal execution probability in the probability storage unit, A work optimizing unit for allocating an operator who performs the above process based on the normal performance probabilities stored in the probability storage unit.

The information on worker abilities stored in the worker information storage section includes an average value (vector) representing the distribution of worker abilities and a covariance (matrix) of the worker ability values. The probability calculation unit calculates a normal performance probability based on the information on the required ability stored in the input information storage unit and the average value and the covariance stored in the worker information storage unit. And

[0010] The input unit inputs information relating to required capabilities of a plurality of types of capabilities for performing the process, and the input information storage unit stores the required capabilities input by the input unit as input information; The information on the worker abilities stored in the worker information storage unit includes an average value representing a distribution of a plurality of types of abilities of the worker, and a covariance of the values of the abilities of the workers. Is characterized in that a normal performance probability is calculated based on the information on the required ability stored in the input information storage unit and the average value and covariance stored in the worker information storage unit.

The input unit inputs the amount of work performed by each process. The input information storage unit stores the amount of work input by the input unit as input information. Is calculated based on the normal performance probability stored in the probability storage unit and the work amount stored in the input information storage unit. The present invention is characterized in that the assignment of workers who perform tasks is optimized based on the overall normal performance probability.

The work optimizing unit calculates a risk, which is a probability that the worker will not normally complete each process, based on the normal execution probability stored in the probability storage unit, and based on the calculated risk, performs the entire work. Is calculated, and the assignment of the worker who performs the work is optimized based on the calculated overall risk of the work.

The worker information storage unit further stores a unit work price of each worker, and the work optimizing unit stores the work unit price stored in the worker information storage unit and the overall normal unit price. The present invention is characterized in that the assignment of the worker who performs the work is optimized based on the performance probability.

[0014] The information on the required capability input at the input unit is a mean value and a covariance representing a distribution of the required capability.

The worker assignment method according to the present invention includes an input step of inputting, as input information, a process name of a process constituting a task to be performed, and information relating to a required capability required as a capability of performing the process. An input information storage step of storing the input information input in the input step in a storage unit, worker names of a plurality of candidate workers for performing the above process, and worker capabilities related to the process execution ability of each worker Worker information storage step of storing information on the storage unit, based on the information on the required capacity stored in the input information storage step and the information on the worker capacity stored in the worker information storage step, Calculate the normal execution probability of each worker successfully completing the above process,
An operation processing step of allocating an operator who performs a process based on the calculated normal performance probability; an output information storage step of storing the operator name of the operator allocated in the operation processing step in a storage unit; An output step of outputting the worker name of the worker stored in the information storage step.

A computer-readable storage medium storing an operator assignment program according to the present invention includes a process name of a process constituting a task to be performed, information on a required capability required as a capability to perform the process, and An input step of inputting the input information as input information, an input information storage step of storing the input information input in the input step in a storage unit, a plurality of workers who are candidates for performing the process, A worker information storing step of storing information on the worker ability related to the worker's process execution ability in the storage unit, and information on the required ability stored in the input information storing step and the worker information storing step. Based on the information on worker abilities, each worker calculates a normal performance probability of successfully completing the above process, and calculates the calculated normal performance probability. An operation processing step of allocating an operator who performs a process based on the probability, an output information storage step of storing an operator name of the operator allocated in the operation processing step in a storage unit, and an output information storage step. An output step of outputting the stored worker name of the worker.

[0017]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, a process model of a virtual enterprise will be described. The reference model of the virtual enterprise has been examined in various fields due to the aforementioned needs. Here, we take the manufacturing industry as an example to the extent necessary for the discussion of risk assessment later, and try to draw a somewhat detailed process model from the following viewpoints. First
Second, the processes within the enterprise are captured in terms of information flow and material flow. Porter (Michael E. Porter et al., "How to Take Evolving Information Technology into Competitive Advantage," DHB, Diamond, November 1985, p.
p. 4-16) considers corporate activities as a “value chain” consisting of “value activities” that create added value.
Value activities are broadly divided into “core activities” and “support activities”. Although the support activities at the porter roughly correspond to the flow of information in the practice of the invention described below, there are differences. For example, product design work is a main activity, but the object to be handled is information, not substances. Second, the ISO CD
14258 (ISO, 1997) hierarchical what,
Complies with the how, do breakdown method. Third
In addition, PDCA (Plan, Do, Check, Acti)
on). W in the above (ISO, 1997)
hat and how are almost equivalent to Plan.
In (1997), the viewpoint of Check and Action is insufficient. Fourth, the activities of the entire enterprise, including the phases of the users of the products, are viewed from the viewpoint of the product life cycle. Fifth, the drawing method is basically IDE
F0 (ICAM definition language)
e) (for example, (Takao Enkawa, Naoki Kunizawa, Junji Tezuka, Moriya Katagiri, Yotaro Miyanishi, "Special Feature CALS", Operations Research, Vol. 40, No. 11, pp. 614-64)
0, 1995)), the flow of information is distinguished by thin lines, and the flow of material is distinguished by thick lines.

The activity of a manufacturing enterprise is basically to envision, design, manufacture and sell products to users. Each activity has an information aspect and a material aspect. As for the input and output in each activity, the information becomes more detailed in the output than the input, that is, the dimension of the information space increases and becomes more specific. Substance is also compared to input
The output will be processed or assembled, increasing the dimensions of the physical space and becoming more specific.
In the case of the information space, the output of the information space is generated by the worker's ability in response to the input. In the case of a physical space, in response to an input, the capabilities of a worker and manufacturing equipment work according to an instruction in the information space, and an output is generated. FIG. 3 shows a process of business activities from suppliers and users from this viewpoint. FIG.
The entire flow of information (not just a simple conversion flow, but a detailed flow of information) and the flow of materials (logistics, which is the movement of simple materials) are also included, but most of the flow of simple movements Rather, the flow of processing and assembling is performed to change to more specific characteristics), but the flow of information and substances also exists hierarchically within one block. FIG. 4 is an exploded view of the “manufacturing” portion of FIG. 3, for example. Other processes can be broken down hierarchically as well.

In FIG. 4, management (planning, monitoring, control)
Plans the execution of the entire block by decomposing the process in charge of this block into smaller sub-processes, monitors whether each sub-process is performed as planned, and deviates from the plan. It is the part that takes appropriate action in cases. PDCA (Pla
n, Do, Check, Action)
It corresponds to the parts P, C and A. The part other than the management part in this block corresponds to the part D in charge of this block.
The manufacturing design cooperates with the planning part of the management to generate the work procedure of each sub-process. That is, the design information provided at the upstream (the specification is “what”, while the design information is “ho”)
The part w) will be further detailed. FIG. 4 shows a process of simply processing and assembling a plurality of materials and parts to produce a single product. In general, the “production” process depends on the type of business, such as “decomposition”, “processing”, The process of “assembly” is a hierarchically configured process. here,
One block consists of information and material, indicating that this block has more or less administrative functions and operates autonomously.

Next, the design process is modeled from the viewpoint of the information space. FIG. 5 shows a model of the design process. In FIG. 5, the input information is I _{d and the} output information is O _d . I _d
Is defined as an n _id dimensional vector, the specification as input information is a certain area in the n _id dimensional information space (for example, the specification speed of a newly designed passenger car is from 140 per hour). 150 km, weight 1000
Kg to 1200 Kg, etc.). The output information O _d as a design result is a _nod dimensional vector, and is a certain area of the _nod dimensional information space. The output information is the target (w
hat) compared to the input information that represents the production method (ho)
information representing the w) is also included, because it is in detail an increase in the number of dimensions, it is n _od> n _id. Also, the output information may be specific, in which case the range of the area is narrow even if it is a variable of the same dimension of the input information (for example, the maximum speed of a newly designed passenger car is 14
5 km). The design process is a process of creating output information from input information by utilizing the capabilities (knowledge, creativity, work speed) of a design worker, including the capabilities of design support equipment. The ability of the design operator (including a design support facilities) that are required to perform the design and R _d.
Since the worker's ability to generate output information is at least the ability to replace the number of dimensions of input information with the number of dimensions of output information, the dimension of R _d is represented by a matrix of n _id × _nod dimensions. And Based on the above premise, the design process can be conceptually expressed by the following equation.

[0021]

(Equation 1)

Next, the manufacturing process is considered and modeled from the viewpoint of the information space and the physical space. FIG. 6 shows a model of the manufacturing process. Input information in FIG. 6 is a design output information O _d, the input material I _m, and the output material O _m. I
Let _{m be} a n _im dimensional vector, n
_This is an area of the _im- dimensional physical space (for example, a width of 2000 m).
m, 10000 mm long, 1 mm thick iron plate). O _m is a _nom- dimensional vector, and is an area of the _nom- dimensional physical space (for example, a door of a passenger car that is processed according to a design drawing).
The output material is processed or assembled is subjected, increases the number of dimensions since become details are n _om> n _im.
Even in the same dimension, the area is a specified area such as a three-dimensional shape processed from a three-dimensional material. The manufacturing process includes the processing and
Demonstrate the ability (skill, work speed) of the assembly worker,
This is a process of producing an output material from an input material as faithfully as possible with design information. The ability of the machining and assembly workers required to accomplish the manufacturing process (including manufacturing facilities) and R _m. An operator's ability to generate an output material is at least the ability to process and assemble the number of dimensions of the input material into the number of dimensions of the output material.
The dimension of _m is assumed to be represented by a matrix of dimension n _im × _nom . The design information O _d is originally input information of the manufacturing process and is drawn as a solid line in FIG. 6. However, since the operator understands this and refers to the information at the time of work, the design information O _d is indicated by a dotted line in FIG. As such, it is considered that it is indirectly included in the ability of the worker. Based on the above assumptions, it is assumed that the manufacturing process can be conceptually expressed by the following equation.

[0023]

(Equation 2)

In both the design and manufacturing processes, the model can be modeled in common with the case where the output information or the output material is generated by the work of the operator on the input information or the input material. The abilities of workers (organizations) are roughly divided into (1) technical abilities (knowledge and creativity), (2) management abilities, and (3) support equipment, and abilities required for a certain process (required abilities). Are refined and specified according to the process. In response to refinement and specification, public qualifications, vendor qualifications, questionnaires, interviews, etc. can also be used as performance indicators.

Embodiment 1 An embodiment showing an example of a worker assignment system, a worker assignment method, and a worker assignment program according to the present invention will be described below based on the virtual enterprise process model described above. FIG. 1 is a block diagram showing an example of this embodiment. Reference numeral 1 denotes an input unit for inputting, as input information, a process name of a process constituting a task to be performed and information on a required capability required as a capability of performing the process. The input unit 1 performs an input step of inputting input information. Reference numeral 2 denotes an input information storage unit that stores input information input by the input unit 1. Further, the input information is stored in the input information storage unit 2 in the input information storage step. Reference numeral 3 denotes a worker information storage unit that stores worker names of a plurality of workers who perform a process and information on worker abilities related to the process execution abilities of each worker. In addition, the worker information storage step stores information on worker abilities in the worker information storage unit 3. 4 indicates that each worker normally completes the above process based on the information on the required ability stored in the input information storage section 2 and the information on the worker ability stored in the worker information storage section 3. An arithmetic processing unit that calculates a probability and assigns an operator who performs a process based on the calculated normal performance probability.
Further, the arithmetic processing unit 4 calculates a normal execution probability and executes an arithmetic processing step of assigning workers. Reference numeral 5 denotes an output information storage unit that stores the worker name of the worker assigned by the arithmetic processing unit 4. Further, the output information storage unit 5 stores the worker name of the worker assigned in the output information storage step. Reference numeral 6 denotes an output unit that outputs the worker name of the worker stored in the output information storage unit 5. Further, the output unit 6 performs an output step of outputting the worker name of the worker.
Reference numeral 7 denotes a probability storage unit that stores the normal execution probability calculated by the arithmetic processing unit 4. Further, the arithmetic processing unit 4 calculates a normal performance probability that the worker completes each process normally,
A probability calculating unit 8 for storing the calculated normal performance probability in the probability storage unit 7, a work optimizing unit 9 for allocating an operator performing the above process based on the normal performance probability stored in the probability storage unit 7; Is included.

The system shown in FIG. 1 is realized on a computer shown as an example in FIG. 201 is an input device, 202 is a storage device, 203 is a control device, 204 is an arithmetic device, 20
5 is an output device. The input unit 1 is included in the input device 201. For example, data may be input via a communication network in addition to input devices such as a keyboard and a mouse. The input information storage unit 2, the worker information storage unit 3, the probability storage unit 7, and the output information storage unit 5 are included in the storage device 202. As an example, a cache memory, a hard disk device, a flexible disk, an optical disk, and the like can be considered. The input information storage unit 2 may be a storage device external to the computer as well as a storage device built in the computer. The arithmetic processing unit 4, the probability calculating unit 8, and the work optimizing unit 9 are included in the arithmetic device 204. The output unit 6 is included in the output device 205, and is, for example, a printer, a screen display device, or the like. The worker assignment program is stored in a computer-readable storage medium, and the program read by the input device 201 operates on the arithmetic device 204.

Next, the input information stored in the input information storage unit 2 will be described. As the input information, at least a process name of a process constituting a task to be performed and information on a required capability required as a capability of performing the process are input from the input unit 1. In this embodiment, a case will be described in which a process name of a process, a required capacity, and a work amount of each process are input information. In the following description, the process name of the input process may be described as a process, but the process name of the process is shown.
Hereinafter, the total number of all processes is represented by I (i = 1,
2,. . . . . , I), each process is denoted by q _i , the required capacity required for each process is denoted by r _i , and the work amount of each process is denoted by w _i . An overall process configuration Q including each process q _i is formulated. Let Q = {q ₁ , q
₂ ,. . . . . , Q _I }. The required capacity r _i for performing each process q _i is determined, and the entire required capacity is defined as
R = {r ₁ , r ₂ ,. . . . . , R _I }. r _i is a vector quantity, which will be described later. Formulated workload w _i of each process q _i, the total amount of work, W = {w _1,
w ₂ ,. . . . . , W _I }. w _i is a scalar quantity determined in an appropriate unit (hour, day, etc.).

Next, the output information stored in the output information storage unit 5 will be described. The output information includes at least the worker name of the worker assigned by the arithmetic processing unit 4. The name of the assigned worker and a risk evaluation value described later may be used as the output information. The worker name and the risk evaluation value are values that can be obtained as output by inputting the input information into the present system. G is the worker name of the worker optimally assigned to each process q _i by this system, or the whole work group or work company. In addition, a risk evaluation value (hereinafter, also referred to as “risk”) is output. Worker names and risks are not only the first optimal value, but also the second, third,
··· Alternatives such as The assigning worker (or working group or working company) has G = ｛g _q1 ,
g _q2,. . . . . , G _qI }. Within this specification,
The worker name of the worker may be described as a worker, but indicates the worker name of the worker. Risk is Risk.

Next, the information stored in the worker information storage unit 3 is stored.
The following describes the internal information. In the worker information storage unit 3
Is at least the work of multiple workers performing the process.
Contractor name and workers related to each worker's process execution ability
Information about the ability is stored. There are other elements
No problem. These pieces of information are input in advance. Weather
Auxiliary workers may be working groups or working companies.
Can be Each worker g _j(The maximum j is J)
Trader's ability information (average value c_j, Covariance Σ_j) The worker information
The information is stored in the information storage unit 3. Average value c_jIs the same as the required capacity
Vector quantity. Covariance Σ_jIs its covariance. C = ｛c₁, C_Two,. . . . . , C_j｝ (5) Σ = ｛Σ₁, Σ_Two,. . . . . , Σ_j｝ (6) C is the average value of each candidate ability c_jIs a set of Σ
Covariance of each candidate abilityΣ_jIs a set of This candidate ability
Information on the force is based on the results after the work
Observe work performance and correct accordingly.

Next, the operation will be described. When assigning a worker to be performed, input information is input from the input unit 1. In the worker assignment system, the probability calculation unit 8 and the work optimization unit 9 of the arithmetic processing unit 4 assign the workers based on the input information and the worker abilities stored in the worker information storage unit 3. The assigned worker is stored in the output information storage unit 5 and output by the output unit 6. The above is the outline of the entire operation.

Hereinafter, the processing of the arithmetic processing unit 4 which is a point of the present invention will be described. In the arithmetic processing unit 4, each worker calculates a normal performance probability of completing the above process normally, calculates a risk based on the calculated normal performance probability, and calculates a risk of the worker performing the process based on the calculated risk. Assign. Therefore, first, a risk evaluation method will be described, and a method of assigning workers will be described. The proposal of the risk evaluation method calculates the risk of a single process, and calculates the risk of the entire combined process based on the calculated single process risk. First, the risks of a single process are described, and then the risks of the entire combined process based on the risks of a single process are described.

The following describes the risks of a single process. In a virtual enterprise, some of the processes are outsourced, which creates uncertainty in the output of the outsourced process. Implementing the process within an organization also creates uncertainty in the output of the process, but because it is within the same organization, it can be resolved by internal efforts, but in the case of a virtual enterprise, it is an obvious problem. It is becoming. As described above, since the output information and the output substance change depending on the ability of the entrusted party, the output information and the output substance are uncertain and distributed due to the uncertainty of the ability. The probability that the output information and output materials will not be as planned is the risk of this process. Since it is difficult to directly obtain the probability distribution of the output of the process itself, the risk is considered based on the relationship between the required ability and the candidate ability. For example, the design capability is expressed as the required capability of the worker.
_d, production capacity and r _m, c _d a design capacity as a candidate ability of candidates, the manufacturing capability and c _m. Because candidates capability there is uncertainty, the average value c _d candidates ability, c _m, the dispersion sigma _d, sigma _m, and those represented by the multidimensional normal distribution. The candidate capability required capacity r _d, the probability of more than r _m, and the probability that the output information of the process, the output material obtained as planned. Here r _d, r _m is the aforementioned required capabilities matrix R _d, it is assumed that rearranged the R _m as a vector. The risk is therefore one minus its probability.

The following is a summary of the definitions of the above risks. The risk can be defined as “the probability that the output result of the process will not be achieved as planned”. In addition, the risk can be paraphrased as “the probability that the assigned worker (organization) does not reach the capability required for the process”. Here, subscript design and production means d,
m is omitted, and the candidate ability x (n-dimensional vector) follows a multidimensional normal distribution of a mean value c (n-dimensional vector) and a covariance matrix Σ (n × n-dimensional matrix) as a probability density distribution function f (x). It is expressed by the following equation.

[0034]

(Equation 3)

As an interpretation of this equation, the average value is the nominal value of the capability, and the variance can be interpreted as indicating the degree of trust in the nominal value. The probability P that the output of the process is normally achieved as planned (hereinafter, referred to as “normal execution probability”) P is the probability that the candidate ability x exceeds the required ability r. P can be calculated by the following equation. Since the candidate ability x and the required ability r are n-dimensional vectors, the following equation is a multiple integration.

[0036]

(Equation 4)

Therefore, the risk Risk is given by the following equation. Risk = 1-P (7)

Next, the risk of the entire work process will be described. In the virtual enterprise, the structure of the entire process is created while considering the contract unit when forming the virtual enterprise. The relationship between each process is patterned into three types of disassembly, assembly, and series as shown in FIG. 3 in FIG.
In the process configuration of the three patterns, a calculation method of the overall normal operation probability (probability that the output as a whole becomes the output as planned) will be considered. If no processes are redundant, as in the calculation method in reliability theory, every process is indispensable for the execution of the whole business,
Therefore, if the execution of each process is an independent event,
It can be considered that the overall normal operation probability is defined as the product (simultaneous probability) of the normal operation probability of each process. But,
This concept is not appropriate in this embodiment. In other words, when the output results of each process are aggregated into an overall output result as in the case of this embodiment, the probability is determined only by whether the process is normal or abnormal as in the reliability calculation method. Is not appropriate.
Further, the work result of a certain process also depends on the work result of the preceding stage, but consideration of the dependency of the work between the processes is a future subject, and here, the dependency is not considered. In this embodiment, the overall normal execution probability is defined as an average of the total amount of work that is normally completed. The calculation procedure of the overall normal performance probability and the overall risk is shown below.

First, the whole process configuration q _i (i = 1,
2,. . . . . , I). The total number of processes is I
And Q = {q ₁ , q ₂ ,. . . . . , Q _I } (8) Next, each process q _i (i = 1, 2,..., I)
The required capacity r _i for performing the above is determined. R = {r ₁ , r ₂ ,. . . . . , R _I } (9) The work amount w _i of each process q _i (i = 1, 2,..., I) is determined. W = {w ₁ , w ₂ ,. . . . . , W _I ｝ (10)

[0040]

(Equation 5)

[0041]

(Equation 6)

[0042]

(Equation 7)

The normal execution probability P of the entire process is defined as a ratio of the average normal execution output amount to the total work amount.

[0044]

(Equation 8)

That is, the overall normal performance probability is a weighted average value of the normal performance probabilities of each process. The risk Risk of the entire process is represented by the following equation.

[0046]

(Equation 9)

That is, the overall risk is a weighted average value of the risk of each process.

Next, an evaluation method for the risk calculated above will be considered. First, in order to explain the risk evaluation method, an example of risk calculation in a certain assumed case will be described. Next, as an application of the risk evaluation method, a risk evaluation when a virtual enterprise is established, that is, when a contract for outsourcing a process is made will be described. Next, we discuss the risks during the process, and finally, make recommendations on the information infrastructure required to realize such functions.

First, a calculation example of the risk evaluation will be described. For simplicity, the candidate abilities are one-dimensional normal distributions. 8 shows a probability density distribution of a candidate's ability, and FIG. 9 shows a probability distribution in which the candidate ability does not reach the required ability, that is, a risk probability distribution. 8 and 9, the candidates (organizations) A, B, and C have average values of 1.5, 1.5, 2..
0 and a normal distribution with standard deviations of 0.25, 0.5 and 1.0. In FIG. 9, the required capacity is a (= 1.0
0), the risk is smaller in the order of candidates A, B, and C. When the required capacity is between a and b (≒ 1.33), the risk is smaller in the order of candidates A, C, and B. small. In general, the risk is planned to be significantly smaller than 1, so, for example, in the example of FIG. 7, the nominal ability (ie, average) of candidate C is 2 and the nominal ability of candidate A is 2 1
This indicates that the risk may be reversed depending on the degree of credit (ie, the variance) even if it exceeds the value.

Next, a description will be given of a risk evaluation when a virtual enterprise is established. When establishing a virtual enterprise, the above evaluation method is used to evaluate the risk of the outsourcing process. In-house processes can also be evaluated in the same way, whether internal procurement or external procurement (ma
Ke or buy) can be determined based on clear criteria. There are two ways of thinking when establishing a virtual enterprise: cost priority and risk priority. Cost priority means allocating the total cost to each process for a certain total cost, evaluating the cost and risk of each process, and selecting candidates (including company) to minimize the risk. To tell. In the case of cost priority, it is desirable to allocate a cost that matches the risk among the total costs as a cost for recovery. Risk priority means selecting candidates so that the risk in all processes is below a certain value. Estimate the cost of performing the process with the selected candidate.

Next, the risk during the operation of the virtual enterprise will be described. Plan the risks and costs of all processes at the time of establishment, and monitor the progress of performance by timely monitoring. Processes that have been completed as planned
At that point the risk disappears. Processes whose work has not been completed as planned take additional measures to carry out the process. Monitoring of progress is not performed for the first time after the entire process has been completed, but it is necessary to activate the monitoring function at the start, end, and some intermediate points at least for each subprocess as an element. Necessary for smooth execution. Generally, the risk decreases as the process progresses. The remaining risks after confirming that the process has completed successfully from 1 to k are expressed as follows (assuming the process set is an all-ordered set).

[0052]

(Equation 10)

Finally, the information infrastructure will be described. In particular, the handling of company capability information will be described. Regarding the development of a social information infrastructure, the following measures should be developed as a social information infrastructure in order to reduce the search cost when establishing a virtual enterprise. (B) Set up a candidate company registration system. (B) Setting public performance indicators. (C) Development of a candidate company information database. (D) Establishment of an information maintenance organization (update of reference models, terminology, capability indicators, maintenance of databases).

It is desired that the transmitting side transmits the following information to the above social information infrastructure. (E) Notify the company's capabilities (organization, employees) on the network based on public capability indicators. (H) Advertise on the network the capabilities of the company other than the public performance indicators. (G) Publicize performance information.

On the receiving side (virtual enterprise coordinating side), it is desirable to obtain information on candidate companies from the social information infrastructure and perform the following unique handling of each company. (H) Set the company's ability index with reference to the announced corporate ability (public index, non-public index). Generally, the received performance index is broken down and the number of dimensions is increased. (I) The performance information indicates the degree of trust in the announced capability and is reflected in the value of the variance matrix described above. (V) Do not rely on the information you receive, but establish your own evaluation method.

As described above, in order for the corporate form of the virtual enterprise to become truly universal, there are still many matters to be prepared as a society as a whole and matters to be prepared independently by companies. One of the most important factors when creating a virtual enterprise is "I'm really happy to outsource my work to an external organization (quality,
Is the construction period, cost) from the viewpoint of risk evaluation, and proposed a risk evaluation method. According to this method, it can be expected that a rational judgment will replace a vague judgment in the past. Further, by reflecting the results, more accurate judgment can be made. The following issues can be considered as future issues. -Method of setting indices (scales) of required ability and candidate ability. -A method of maintaining such capability information as a social infrastructure. -A method of estimating and learning parameters based on the ability and performance declared by the sender using any algorithm, and providing it as information on public corporate ability. -A unique utilization method of such information as a company (refinement of the capability dimension of the company, setting of a covariance matrix).・ Risk evaluation method when there is a dependency between processes.

Embodiment 2 In the first embodiment, the method of evaluating the risk has been described. In the present embodiment, a case of evaluating the cost and the risk will be described. In this embodiment, as shown in FIG. 10, the worker unit price of the candidate worker is stored in the worker information storage unit 3a. Similarly to the risk, the cost of the entire work can be calculated from the work amount and the work unit price by storing the work unit price of the candidate worker as internal information in the worker information storage unit 3a. The calculated cost and risk of the entire work can be provided as judgment material for worker assignment. By selecting the assignment of the worker based on these judgment factors, the assignment of the worker can be optimized.

Embodiment 3 FIG. In the worker assignment system according to this embodiment, a worker j is added to a process i (total number I).
An optimal solution (evaluation criterion of minimum risk) when allocating (total number J) is obtained, but the present invention does not propose an algorithm for this. The point of the present invention is that the risk in a certain assignment pattern can be calculated. Therefore, the optimal algorithm can determine the optimal assignment of the workers by brute-force checking the combination pattern of the process i and the worker j, but the number thereof becomes enormous (combination _J C _I ). Therefore, by classifying the candidates, or by classifying the process according to the difficulty level (degree of required ability),
A device for reducing the number of combinations can be devised. By classifying such candidates, it is possible to efficiently determine the optimal assignment of workers.

Embodiment 4 FIG. In the above-described embodiment, the case where the dimensions of the vector of the required ability of the worker and the vector of the candidate ability are different will be described. The required ability and the candidate ability need not be vectors of the same dimension. However, there is no problem when the element included in the dimension of the required ability vector is included in the candidate ability vector. Only the elements of the candidate capability vector that are not in the required capability are ignored. More specifically, if the required capability requires the design capability, and if the candidate capability includes the design capability and the manufacturing capability, the design capability is used, but the manufacturing capability is ignored. On the other hand, when the element included in the required ability vector is not in the candidate ability vector, the following measures are considered necessary. For a candidate having a vector that is not in the candidate ability in the required ability, a new input request is made for a vector that is not in the candidate ability, or a candidate that has a vector that is in the required ability and is not in the candidate ability is: It is necessary to take measures to exclude the assignment of the worker. Specifically, if the required capacity requires design capacity and manufacturing capacity,
If the candidate ability has no design ability and has a manufacturing ability, the input of the design ability is requested or the candidate is excluded from the assignment of the worker. In this way, even when the required ability and the candidate ability have different vectors, the optimization of the worker can be performed.

Embodiment 5 FIG. In the above embodiment, the case where the required ability and the candidate ability are multidimensional has been described.
The required ability and the candidate ability may be a one-dimensional case. In this case, too, the average value and the variance can be calculated from the one-dimensional vector candidate ability to optimize the worker assignment. The average value (average value vector) in multiple dimensions corresponds to the average value in one dimension. The average value is a vector value in multiple dimensions, but is a scalar value in one dimension. Covariance in multiple dimensions (covariance matrix) corresponds to variance in one dimension. Further, as in the fourth embodiment, even if the required ability is a one-dimensional vector, the candidate ability may be a multidimensional vector.

Embodiment 6 FIG. Further, even when the required capacity is a random variable, the present invention can be implemented by the worker assignment system according to the present invention, and the worker assignment method can be used. FIG. 11 shows an example. The average value r _i and the covariance _r Σ _i of the required capacity are input from the input unit 1 to the input information storage unit 2b as input information. This case can be realized by using convolution integration.

Embodiment 7 In the above embodiment, the case where the entire work is composed of a plurality of processes has been described. However, even when the number of processes is one, the operation can be performed by the worker assignment system according to the present invention. ,Also,
A worker assignment method can be used.

Embodiment 8 FIG. There are two types of work performance, demand ability and candidate ability, both of which are evaluated on a common scale. The evaluation criteria for abilities change with the times, such as the emergence of new technologies. Here, evaluation criteria are evaluated using multi-dimensional vectors, but with the advent of new technologies, there are also evaluation criteria for additional dimensions, and even in the same dimension, the general world In some cases, the standard increases. That is, when the present system is applied, a standard such as a capability standard for a certain year is set. Review this, for example, every year, add dimensions,
Delete and update each candidate's ability table. In the above embodiment, the probability variables, the case is a normal distribution, even if it is any other distribution such exponential distribution and chi ² distribution, the numerical expressions corresponding to (6) By integrating, it is possible to apply this worker assignment system and method.

[0064]

According to the worker allocating system, the worker allocating method and the computer readable storage medium storing the worker allocating program according to the present invention, the worker optimized based on the normal execution probability of the work. You can get the assignment.

Further, according to the worker assignment system according to the present invention, it is possible to obtain an optimized worker assignment based on the normal execution probability of the work calculated in consideration of the average value and the dispersion of the worker abilities. .

Further, according to the worker assignment system according to the present invention, the worker assignment optimized based on the normal execution probability of the work calculated in consideration of the average value of the multidimensional worker ability and the covariance is performed. Obtainable.

According to the worker assignment system according to the present invention, it is possible to obtain an optimized worker assignment based on the normal execution probability of the entire work calculated in consideration of the work amount of each process.

According to the worker assignment system according to the present invention, it is possible to calculate a risk from the normal execution probability of the work and obtain an optimized worker assignment based on the risk of the entire work calculated from the risk.

Further, according to the worker assignment system according to the present invention, it is possible to obtain an optimized worker assignment based on costs.

Further, according to the worker assignment system according to the present invention, an optimized worker assignment can be obtained even when the required ability required of the worker is an average value representing the distribution of the ability.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an example of Embodiment 1 of the present invention.

FIG. 2 is a diagram illustrating an example of a basic block of a computer.

FIG. 3 is a diagram illustrating an example of a process model of a company activity.

FIG. 4 is a detailed view of a manufacturing portion in the business activity process model of FIG. 3;

FIG. 5 is a diagram showing a model of a design process.

FIG. 6 is a diagram showing a model of a manufacturing process.

FIG. 7 is a diagram showing a pattern of an overall process configuration.

FIG. 8 is a diagram illustrating an example of a probability density distribution of candidate abilities.

FIG. 9 is a diagram illustrating an example of a probability distribution of a risk with respect to a required capacity.

FIG. 10 is a block diagram illustrating an example of Embodiment 2 of the present invention.

FIG. 11 is a block diagram illustrating an example of Embodiment 6 of the present invention.

[Explanation of symbols]

1 input section, 2, 2b input information storage section, 3, 3a worker information storage section, 4 arithmetic processing section, 5 output information storage section,
6 output unit, 7 probability storage unit, 8 probability calculation unit, 9 work optimization unit, 201 input device, 202 storage device, 2
03 control unit, 204 arithmetic unit, 205 output unit.

Claims (10)

[Claims] 1. An input unit for inputting, as input information, a process name of a process constituting a task to be performed and information on a required capability required as a capability of performing the process, and an input input by the input unit An input information storage unit for storing information; a worker information storage unit for storing worker names of a plurality of workers who are candidates for performing the above process; Based on the information on the required ability stored in the input information storage section and the information on the worker ability stored in the worker information storage section, a normal performance probability that each worker completes the process normally. And an arithmetic processing unit for allocating an operator who performs a process based on the calculated normal execution probability, and an operator name of the operator allocated by the arithmetic processing unit. Worker assignment system comprising: the output information storage unit that 憶, and an output unit for outputting operator name stored worker to the output information storage unit. 2. The worker assignment system further includes a probability storage unit that stores the normal execution probability. The input unit relates to process names of a plurality of processes and a plurality of required capabilities corresponding to the processes. And input information including information, the arithmetic processing unit calculates a normal performance probability that the worker completes each process normally, and a probability calculation unit that stores the calculated normal performance probability in the probability storage unit. 2. The worker assignment system according to claim 1, further comprising: a work optimization unit that assigns a worker who performs the process based on the normal execution probability stored in the probability storage unit. 3. The information on worker abilities stored in the worker information storage unit includes an average value (vector) representing a distribution of worker abilities and a covariance (matrix) of worker abilities values. The probability calculation unit calculates a normal performance probability based on the information on the required ability stored in the input information storage unit and the average value and covariance stored in the worker information storage unit. 3. The worker assignment system according to claim 2, wherein: 4. The input unit inputs information relating to required capabilities of a plurality of types of capabilities for performing the process, and the input information storage unit stores the required capabilities input by the input unit as input information. The information on the worker abilities stored in the worker information storage unit includes an average value representing a distribution of a plurality of types of abilities of the worker, and a covariance of the values of the abilities of the worker, and the probability The calculation unit calculates a normal performance probability based on the information on the required ability stored in the input information storage unit and the average value and covariance stored in the worker information storage unit. Item 2. The worker assignment system according to Item 2. 5. The input unit inputs the amount of work performed by each process. The input information storage unit stores the amount of work input by the input unit as input information. The conversion unit calculates a total normal performance probability that is a probability that the entire work is normally completed based on the normal performance probability stored in the probability storage unit and the work amount stored in the input information storage unit, The worker assignment system according to any one of claims 2 to 4, wherein the assignment of the worker who performs the work is optimized based on the calculated overall normal execution probability. 6. The work optimizing unit calculates a risk, which is a probability that the worker will not normally complete each process, based on the normal performance probability stored in the probability storage unit, and based on the calculated risk. 6. The worker assignment system according to claim 5, wherein a risk of the entire work is calculated, and an assignment of a worker performing the work is optimized based on the calculated risk of the whole work. 7. The worker information storage unit further stores a work unit price of each worker per unit, and the work optimization unit stores the work unit price stored in the worker information storage unit, The worker assignment system according to claim 5, wherein the assignment of the worker who performs the task is optimized based on the overall normal execution probability. 8. The worker according to claim 1, wherein the information on the required capability input at the input unit is a mean value and a covariance representing a distribution of the required capability. Assignment system. 9. An inputting step of inputting, as input information, a process name of a process constituting a task to be performed and information relating to a required capability required as a capability of performing the process, and an input inputting in the inputting process. An input information storage step of storing information in a storage unit, worker names of a plurality of candidates for performing the above-described process, and information on worker abilities related to process execution abilities of each worker are stored in the storage unit. A worker information storing step, and information on the required ability stored in the input information storing step and information on the worker ability stored in the worker information storing step. An operation processing step of calculating a normal execution probability to be completed, and assigning an operator who performs a process based on the calculated normal execution probability; An output information storage step of storing the worker name of the worker assigned in the storage unit in the storage unit, and an output step of outputting the worker name of the worker stored in the output information storage step. Worker assignment method to be performed. 10. An inputting step of inputting, as input information, a process name of a process constituting a task to be performed and information relating to a required capability required as a capability of performing the process, and an inputting input in the inputting process. An input information storage step of storing information in a storage unit, worker names of a plurality of workers who are candidates for performing the above-described process, and information on worker abilities related to process execution abilities of each worker are stored in the storage unit. A worker information storing step, and information on the required ability stored in the input information storing step and information on the worker ability stored in the worker information storing step. An operation processing step of calculating a normal execution probability to be completed in step (c), and assigning an operator who performs a process based on the calculated normal execution probability; A worker assignment program comprising: an output information storage step of storing a worker name of a worker assigned in a process in a storage unit; and an output step of outputting the worker name of the worker stored in the output information storage step. A computer-readable storage medium on which is recorded.