WO2018163769A1

WO2018163769A1 - Information processing system, information processing method, and computer-readable recording medium

Info

Publication number: WO2018163769A1
Application number: PCT/JP2018/005675
Authority: WO
Inventors: Jun Kawada
Original assignee: Ricoh Co Ltd
Current assignee: Ricoh Co Ltd
Priority date: 2017-03-10
Filing date: 2018-02-19
Publication date: 2018-09-13
Anticipated expiration: 2019-09-10

Abstract

According to an embodiment, an information processing system includes an estimating unit and a selecting unit. The estimating unit is configured to estimate correspondence information representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio that is a ratio at which inspections are omitted based on operation data indicating task-related information. The selecting unit is configured to select one of pairs of the execution error incidence ratio and the corresponding non-inspection ratio included in the correspondence information.

Description

INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER-READABLE RECORDING MEDIUM

The present invention relates to an information processing system, an information processing method, and a computer-readable recording medium.

In the logistics workplace, many staff-hours are spent in inspecting process for picking work. To reduce the staff-hours, an inspection assisting technology using a handy terminal and an inspection automating technology using image recognition or weighing have been known. Also known from the viewpoint of task management is a technology for identifying and assessing tasks, such as inspections, assigned to workers who perform the tasks (for example, see Patent Document 1).

However, it is difficult to deploy such an inspection assisting technology or an inspection automating technology, because such a technology requires a large-scale investment, and imposes a limitation that such a technology makes it difficult to implement a change in response to a change in the business climate or products handled. As a result, many tasks remain dependent on human interventions, and the picking work has the problem that the picked products need to be inspected to detect human errors that occur at a probability of less than 0.1 percent in the entire picking work, including 99.9 percent or more of the products in which no errors are found.

In view of the above problem of the conventional art, there is a need to provide an information processing system, an information processing method, and a computer-readable recording medium having a program for allowing manual inspections to be performed efficiently.

According to the embodiment, manual inspections can be performed efficiently.

Fig. 1 is a schematic illustrating an example of a general configuration of an information processing system according to an embodiment. Fig. 2 is a schematic illustrating an example of a hardware configuration of an information processing apparatus according to the embodiment. Fig. 3 is a schematic illustrating an example of functions provided to the information processing apparatus according to the embodiment. Fig. 4 is a schematic illustrating an example of task information according to the embodiment. Fig. 5 is a conceptual schematic illustrating a model for estimating an execution error incidence ratio and a non-inspection ratio, for each of a plurality of input parameters according to the embodiment. Fig. 6 is a schematic illustrating an example of an estimation graph. Fig. 7 is a schematic illustrating an example of a distribution of error ratios corresponding to respective workers. Fig. 8 is a flowchart illustrating an exemplary operation performed by the information processing apparatus according to the embodiment. The accompanying drawings are intended to depict exemplary embodiments of the present invention and should not be interpreted to limit the scope thereof. Identical or similar reference numerals designate identical or similar components throughout the various drawings.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present invention.
As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
In describing preferred embodiments illustrated in the drawings, specific terminology may be employed for the sake of clarity. However, the disclosure of this patent specification is not intended to be limited to the specific terminology so selected, and it is to be understood that each specific element includes all technical equivalents that have the same function, operate in a similar manner, and achieve a similar result.
One embodiment of an information processing system, an information processing method, and a computer-readable recording medium having a program according to the present invention will now be explained in detail with reference to the appended drawings.

Fig. 1 is a schematic illustrating an example of a general configuration of an information processing system 1 according to the embodiment. As illustrated in Fig. 1, the information processing system 1 includes a server 10 and an information processing apparatus 20, and the server 10 and the information processing apparatus 20 are enabled to connect to each other via a network 30, such as the Internet.

The server 10 stores therein operation data that is based on data collected from an external enterprise system (such as a warehouse management system (WMS)) or from sensor devices attached to workers. The operation data is information related to tasks. In this example, a task is a picking task. Specific content of the operation data will be described later. In the example illustrated in Fig. 1, the server 10 includes an operation data database (hereinafter, referred to as an "operation data DB") 11 for storing therein the operation data.

A configuration of the information processing apparatus 20 will now be explained. In this example, the information processing apparatus 20 is configured as a computer such as a client computer. Fig. 2 is a schematic illustrating an example of a hardware configuration of the information processing apparatus 20 according to the embodiment. As illustrated in Fig. 2, the information processing apparatus 20 includes a central processing unit (CPU) 101, a read-only memory (ROM) 102, a random access memory (RAM) 103, an input unit 104, a display unit 105, and a communication interface (I/F) 106.

The CPU 101 comprehensively controls operations of the information processing apparatus 20. The CPU 101 implements various functions provided to the information processing apparatus 20 by executing various control programs stored in the ROM 102, using a predetermined area of the RAM 103 as a working area. Specific details of the functions provided to the information processing apparatus 20 will be described later.

The ROM 102 is a nonvolatile memory (non-rewritable memory) for storing therein a computer program or various types of setting information related to the information processing apparatus 20.

The RAM 103 is a storage unit such as a synchronous dynamic random access memory (SDRAM), and serves as a working area of the CPU 101 and plays a role as a buffer, for example.

The input unit 104 is a device for receiving an operation of a user.

The display unit 105 is a device for displaying various types of information related to the information processing apparatus 20, and is configured as a liquid crystal display, for example. The input unit 104 and the display unit 105 may also be integrally configured (configured as a touch panel, for example).

The communication I/F 106 is an interface for connecting to the network 30.

Fig. 3 is a schematic illustrating an example of the functions provided to the information processing apparatus 20 according to the embodiment. As illustrated in Fig. 3, the information processing apparatus 20 includes an acquiring unit 201, an estimating unit 202, a display control unit 203, a selecting unit 204, a determining unit 205, and a switching unit 206. For the convenience of the explanation, mainly illustrated in the example illustrated in Fig. 3 are the functions related to the embodiment, but the functions provided to the information processing apparatus 20 are not limited thereto.

The acquiring unit 201 acquires the operation data from the server 10. The operation data is information in which at least worker information, a product (object) type, and a product (object) quantity are associated with each other. The information is, however, not limited thereto, and the operation data may also include a location indicating a work place, work time, an inspection result, and a size and a weight of the product (object), for example, and such data may be reflected to data processing, which will be described later. Furthermore, an execution error in this example is a picking error. In this embodiment, the operation data includes task history data illustrated in a section (A) of Fig. 4, inspection history data illustrated in a section (B) of Fig. 4, product packaging data illustrated in a section (C) of Fig. 4, worker data illustrated in a section (D) of Fig. 4, and worker positioning data illustrated in a section (E) of Fig. 4, but is not limited thereto.

As illustrated in the section (A) of Fig. 4, the task history data is information in which a worker ID for identifying a worker, a foldable container ID for identifying a foldable container, a Japanese article number (JAN) for identifying a product, a quantity, a location, task start time, and task completion time are associated with each other.

As illustrated in the section (B) of Fig. 4, the inspection history data is information in which a worker ID, a foldable container ID, a JAN, a quantity, an inspection result, a cause of an error, an inspector ID for identifying an inspector, and inspection time are associated with each other.

As illustrated in the section (C) of Fig. 4, the product packaging data is information in which a JAN, a product name, a quantity per container, a depth size, a width size, a height size, and a weight are associated with each other. In this example, the size of a product is indicated as a volume, and is obtained by multiplication of a depth size, a width size, and a height size.

As illustrated in the section (D) of Fig. 4, the worker data according to the embodiment is information in which a worker ID, a name, a hire date, and cumulative working hours are associated with each other. The worker data (worker information) may be any information capable of identifying a worker, and may enable a worker to be identified with a worker ID and include information such as the cumulative working hours, for example.

As illustrated in the section (E) of Fig. 4, the worker positioning data is information in which a worker ID, a location, and time associated with each other.

The explanation will now be continued referring back to Fig. 3. The estimating unit 202 estimates correspondence information representing a corresponding relation between an execution error incidence ratio, and a non-inspection ratio indicating a ratio of articles for which inspections are omitted (a ratio with respect to the entire articles) (a ratio of articles not subjected to inspections (inspection-less ratio)) based on the operation data acquired by the acquiring unit 201. The correspondence information in this example is a graph representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio, but the correspondence information is not limited thereto, and may be configured as a table, for example. The estimating unit 202 can estimate the execution error incidence ratio and the non-inspection ratio based on the operation data using Bayesian inference, for each of a plurality of input parameters. Furthermore, the estimating unit 202 can also estimate the execution error incidence ratio and the non-inspection ratio, for each of a plurality of input parameters, using a learning model that is built based on the operation data, for example. The learning model is a model indicating a relation of the execution error incidence ratio and the non-inspection ratio, with respect to an input parameter, for example. The learning model may be configured to, when an input parameter is input to the learning model, output the execution error incidence ratio and the non-inspection ratio corresponding to the input parameter. Various known technologies may be used as a method for building such a learning model.

In this example, as an example, the input parameters are a set of worker information (worker ID), cumulative work time, a product, a product quantity, a product size, and a location. The input parameters may be at least a combination of worker information, an object type, and an object quantity.

Fig. 5 is a conceptual schematic illustrating a model for estimating the execution error incidence ratio and the non-inspection ratio for each of the input parameters. The example illustrated in Fig. 5 indicates that the estimation result corresponding to one input parameter that consists of a worker ID [W1], cumulative working hours [T1], a JAN [J1], a quantity [C1], a size [S1], and a location [L1] (which may be considered as operational conditions for one picking task) is [R1]. Furthermore, the example illustrated in Fig. 5 also indicates that the estimation result corresponding to one input parameter that consists of a combination of the worker ID [W1], the cumulative working hours [T1], a JAN [J2], a quantity [C2], a size [S2], and a location [L2] is [R2]. The example illustrated in Fig. 5 also indicates that an estimation result corresponding to one input parameter consisting of a combination of the worker ID [W1], the cumulative working hours [T1], the JAN [J2], a quantity [C3], the size [S2], and the location [L2] is [R3].

The estimating unit 202 then plots a pair [P1] of the error incidence ratio and the non-inspection ratio corresponding to the estimation result [R1], a pair [P2] of the error incidence ratio and the non-inspection ratio corresponding to the estimation result [R2], and a pair [P3] of the error incidence ratio and the non-inspection ratio corresponding to the estimation result [R3], successively, as illustrated in Fig. 6. By plotting the estimation result for each of the input parameters, it is possible to estimate a graph representing a corresponding relation with the non-inspection ratio. In the explanation hereunder, this graph is sometimes referred to an "estimation graph".

Explained now is an exemplary method for acquiring an execution error incidence ratio using the Bayesian inference. Denoting an incidence of an error as x=1, and denoting a no-error incidence as x=0 in one picking task, a probability distribution can be represented as a Bernoulli distribution expressed by the following Equation (1), using a probability p of an occurrence of an error (execution error incidence ratio).

Because it is likely that the probability p differs depending on the individuals, the probability p is acquired using Bayesian inference for each person. The prior distribution in the Bayesian inference is set as a β distribution (conjugate prior distribution), for the convenience in the calculation, and the β distribution can be expressed as the following Equation (2).

By multiplying this prior distribution with a likelihood acquired from the Bernoulli distribution, a posterior distribution expressed in the following Equation (3) is obtained.

In this example, the posterior distribution is calculated for each picking task, and the calculated posterior distribution is updated as a value in the prior distribution in the next task. When the Bayesian inference is used as a method for acquiring the probability p of an occurrence of an error, p(x=1|D) in the following Equation (4) provides the probability of x being 1 (making an error), provided that data D is given as a condition (data D corresponding to the execution conditions), and a predictive distribution can be acquired by marginalizing the distribution of x with respect to p.

To calculate Equation (4), the following Equation (5) is acquired.

By applying the error ratio distribution corresponding each worker to the β distribution using the above-described Equation (5), a distribution such as that illustrated in Fig. 7 is estimated, for example, and the probability p can be estimated by applying α and β that are acquired from this result, to the prior distribution.

The explanation will now be continued referring back to Fig. 3. The display control unit 203 performs control to display the estimation graph on the display unit 105. An administrator performs an operation of selecting a point based on the operational condition, from the points plotted in the estimation graph (defined by the pairs of an execution error incidence ratio and a non-inspection ratio) (the operation may be a touching operation, or bringing a cursor to the point and clinking on the point, for example).

The selecting unit 204 selects one of the pairs of the execution error incidence ratio and the corresponding non-inspection ratio included in the correspondence information. In this example, the selecting unit 204 selects one point (corresponding to one "pair") from the points in the estimation graph, in response to an operation of an administrator. The selection is, however, not limited thereto, and an optimal value may be automatically selected through machine learning, for example.

The determining unit 205 determines a condition for determining whether an inspection is to be performed based on the pair selected by the selecting unit 204. In this example, the determining unit 205 determines the determining condition whether an inspection is to be performed using the point selected from the estimation graph by the selecting unit 204 as a boundary. For example, using the input parameter corresponding to the point selected by the selecting unit 204 in the estimation graph as a reference point, the determining unit 205 can establish (set) a picking task condition under which an execution error incidence ratio is higher than and a non-inspection ratio is lower than those at the reference point in the estimation graph, as a determining condition for performing an inspection. In other words, using one point in the estimation graph as a reference point, the determining condition can be established in such a manner that inspections are to be performed for objects with an error incidence ratio higher than that of the reference point and a non-inspection ratio lower than that of the reference point, and no inspection can be performed for objects with an error incidence ratio lower than that of the reference point and a non-inspection ratio higher than that of the reference point. The switching unit 206, which will be described later, then performs switching control to perform an inspection when the operational condition of a picking task (corresponding to an input parameter) matches the thus established determining condition, and not to perform an inspection when the operational condition does not match the determining condition.

The switching unit 206 performs switching control to switch whether an inspection is to be performed based on the determining condition established by the determining unit 205 in advance. For example, the switching unit 206 may be configured to perform control, when switching is made to perform the inspection, to notify a handy terminal (terminal) held by the inspector that an inspection is to be performed. The switching unit 206 may also be configured to perform control, when switching is made not to perform the inspection, to notify the handy terminal (terminal) held by the inspector that no inspection is to be performed, for example.

Fig. 8 is a flowchart illustrating an exemplary operation performed by the information processing apparatus 20 when the determining condition is determined. Because the specific operation performed at each step is as described above, detailed explanations thereof will be omitted as appropriate. As illustrated in Fig. 8, the acquiring unit 201 acquires the operation data (Step S1). The estimating unit 202 then estimates a graph representing a corresponding relation between the execution error incidence ratio and the non-inspection ratio (estimation graph) based on the operation data acquired at Step S1 (Step S2). The selecting unit 204 then selects one of the points in the estimation graph, in response to an operation performed by the administrator (Step S3). The determining unit 205 then determines the determining condition for determining whether an inspection is to be performed, using the point selected in the estimation graph at Step S3 as a boundary (Step S4).

As explained above, in this embodiment, a graph representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio is estimated based on the operation data indicating task-related information, and a determining condition for determining whether an inspection is to be performed is determined using a point selected from the graph, selected in response to an operation of the administrator, as a boundary. In this manner, it is no longer necessary to perform an inspection for every task, so that the inspection can be performed efficiently without deploying any automating technology for the inspection. Furthermore, in this embodiment, a picking task is used as an example of the task, but the task is not limited thereto, and the present invention may be used prior to an inspection in any other tasks such as product storing tasks in the logistic operations, or assembly tasks in the manufacturing operations.

Furthermore, for example, the estimating unit 202 may also estimate the correspondence information representing a corresponding relation between an inspection ratio that indicates a ratio at which the inspections are performed and an execution error incidence ratio.

Although one embodiment of the present invention is explained above, the present invention is not limited to the embodiment as it is, and may also be embodied by modifying some of the elements, within the scope not deviating from the essence thereof, during the implementation stage. Furthermore, various inventions may be formed by combining a plurality of elements disclosed in the embodiment as appropriate. For example, some elements may be omitted from the entire elements explained in embodiment.

For example, the information processing apparatus 20 according to the embodiment may be configured as a group of a plurality of apparatuses, and the functions illustrated in Fig. 3 may be installed in the group of apparatuses in a distributed manner. In summary, the information processing system 1 may have any configuration at least including a function corresponding to the estimating unit 202, a function corresponding to the selecting unit 204, and a function corresponding to the determining unit 205, and may be implemented as one apparatus or as a plurality of apparatuses.

Furthermore, a computer program executed by the information processing system 1 according to the embodiment may be provided in a manner recorded in a computer-readable recording medium such as a compact disc read-only memory (CD-ROM), a flexible disk (FD), a compact disc recordable (CD-R), a digital versatile disc (DVD), and a universal serial bus (USB), as a file in an installable or executable format, or may be provided or distributed over a network such as the Internet. Furthermore, various computer programs may be provided incorporated in a ROM or the like in advance.
The above-described embodiment is illustrative and does not limit the present invention. Thus, numerous additional modifications and variations are possible in light of the above teachings. For example, at least one element of different illustrative and exemplary embodiment herein may be combined with each other or substituted for each other within the scope of this disclosure and appended claims. Further, features of components of the embodiment, such as the number, the position, and the shape are not limited the embodiment and thus may be preferably set. It is therefore to be understood that within the scope of the appended claims, the disclosure of the present invention may be practiced otherwise than as specifically described herein.
Further, any of the above-described apparatus, devices or units can be implemented as a hardware apparatus, such as a special-purpose circuit or device, or as a hardware/software combination, such as a processor executing a software program.
Further, as described above, any one of the above-described and other methods of the present invention may be embodied in the form of a computer program stored in any kind of storage medium. Examples of storage mediums include, but are not limited to, flexible disk, hard disk, optical discs, magneto-optical discs, magnetic tapes, nonvolatile memory, semiconductor memory, read-only-memory (ROM), etc.
Alternatively, any one of the above-described and other methods of the present invention may be implemented by an application specific integrated circuit (ASIC), a digital signal processor (DSP) or a field programmable gate array (FPGA), prepared by interconnecting an appropriate network of conventional component circuits or by a combination thereof with one or more conventional general purpose microprocessors or signal processors programmed accordingly.

1 Information processing system
10 Server
11 Operation data DB
20 Information processing apparatus
30 Network
201 Acquiring unit
202 Estimating unit
203 Display control unit
204 Selecting unit
205 Determining unit
206 Switching unit

Japanese Laid-open Patent Application No. 2008-201569

Claims

　　An information processing system comprising:
　　an estimating unit configured to estimate correspondence information representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio that is a ratio at which inspections are omitted based on operation data indicating task-related information, and
　　a selecting unit configured to select one of pairs of the execution error incidence ratio and the corresponding non-inspection ratio included in the correspondence information.
　　The information processing system according to claim 1, further comprising a determining unit configured to determine a determining condition whether an inspection is to be performed based on the pair selected by the selecting unit.
　　The information processing system according to claim 2, further comprising a switching unit configured to perform control to switch whether an inspection is to be performed based on the determining condition.
　　The information processing system according to any one of claims 1 to 3, wherein
　　the task is a picking task,
　　the execution error is a picking error, and
　　the operation data is information in which at least worker information, an object type, and an object quantity are associated with each other.
　　The information processing system according to claim 4, wherein
　　the estimating unit estimates the execution error incidence ratio and the non-inspection ratio based on the operation data using Bayesian inference, for each of a plurality of input parameters.
　　The information processing system according to claim 4, wherein
　　the estimating unit estimates the execution error incidence ratio and the non-inspection ratio using a learning model built based on the operation data, for each of a plurality of input parameters.
　　The information processing system according to claim 5 or 6, wherein
　　the input parameters are at least a combination of worker information, an object type, and an object quantity.
　　The information processing system according to claim 1, wherein
　　the estimating unit estimates correspondence information representing a corresponding relation between an inspection ratio indicating a ratio at which the inspections are performed, instead of the non-inspection ratio, and the execution error incidence ratio.
　　An information processing method comprising:
　　estimating correspondence information representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio that is a ratio at which inspections are omitted based on operation data indicating task-related information; and
　　selecting one of pairs of the execution error incidence ratio and the corresponding non-inspection ratio included in the correspondence information.
　　A computer-readable recording medium that contains a computer program that causes a computer to execute:
　　estimating correspondence information representing a corresponding relation between an execution error incidence ratio and a non-inspection ratio that is a ratio at which inspections are omitted based on operation data indicating task-related information; and
　　selecting one of pairs of the execution error incidence ratio and the corresponding non-inspection ratio included in the correspondence information.