KR101433139B1 - Data Preprocessing System For Process Analysis - Google Patents

Abstract

A data preprocessing system for analyzing a process according to the present invention includes: a definition module including a workflow definition unit for defining data to be extracted and an SQL definition unit for defining how to query the data in a database; A data extraction module for extracting data from a database using information recorded in the definition module; And a standard conversion module for receiving data from the data extraction module and converting the data into a standard format file for process analysis.

Description

[0001] The present invention relates to a data preprocessing system for process analysis,

The present invention relates to a data processing system. More specifically, to a data preprocessing system that extracts data from a database and provides the data in a standard format for process analysis.

As international trade increases due to globalization, marine transportation using container ships takes up a large part in international trade. In Korea, where three sides are surrounded by sea, most of the international transportation is occupied by maritime transportation.

Currently, many countries including Korea are developing various automated container terminal systems to enhance the efficiency of ports that serve as a base for marine transportation.

The Terminal Operation System (TOS) receives information about containers from shipping companies, shipping companies, ships entering the port, and manages the flow of containers in the terminal so that all containers can be moved in the most efficient way to the correct location. System, which is an essential element for automation of container terminals.

Because TOS stores various kinds of information for managing the container flow and various event information about the container in the database, analyzing the event log recorded in the TOS using the process mining program can grasp the current container management process, Through the identified process, a new process can be designed to detect problems in the current system and to remedy the problems.

However, in order to convert the event log recorded in the TOS into the MXML (mining XML) format required by the process mining program, a specialist who understands all the storage methods of the database is required, and the extraction method is also very cumbersome. There is a problem in that the entire process for extraction must be performed again.

An object of the present invention is to provide a data preprocessing system for extracting data to be used for process mining from a database.

Another object of the present invention is to provide a data preprocessing system for extracting data to be used for process mining from a database by a user having no expert knowledge on the TOS database storing method by having a workflow defining unit and an SQL defining unit .

It is still another object of the present invention to provide a data preprocessing system capable of automatically reusing an SQL statement by having a workflow definition unit and an SQL definition unit.

Yet another object of the present invention is to provide a data preprocessing system capable of extracting a standard format file for process analysis directly from a database without having to generate an intermediate file by providing a data extraction module and a standard conversion module.

It is still another object of the present invention to provide a system and method for creating / modifying / deleting a workflow definition part and an SQL definition part, checking / changing an SQL statement corresponding to a workflow definition part scheme, To provide a data preprocessing system that can provide the data.

In order to accomplish the above object, a data preprocessing system for analyzing a process according to an embodiment of the present invention includes a workflow definition unit for defining data to be extracted and an SQL definition unit for defining how to query the data in a database Definition module; A data extraction module for extracting data from a database using information recorded in the definition module; And a standard conversion module for receiving data from the data extraction module and converting the data into a standard format file for process analysis.

The workflow definition includes a scheme for inquiring an object related to a process and a scheme for inquiring an event related to a target. Each scheme includes an identifier (ID) for searching an SQL statement corresponding to each scheme from the SQL definition unit, .

Unlike schemes for querying an object, schemes for querying an event additionally require information of a workflow model element and an event type (Event Type) indicating which action the event is for.

The SQL statement includes a variable for querying the ID of the target. The SQL statement for querying the event includes a variable for inquiring the ID of the related object and the time at which the event occurred.

Each SQL statement contains a data query time span in the conditional statement, and the query time span is the same for all SQL statements when the user enters the desired data query time span.

The data preprocessing system according to another embodiment of the present invention may further include a definition creation module that provides a user interface that allows the user to create, modify, and delete data stored in the workflow definition unit or the SQL definition unit.

The user interface may provide SQL definition display, modification, and testing functions corresponding to the schemes, and may also provide a function to select or exclude specific schemes when extracting data.

The present invention can easily extract data to be used for process mining from a database even if a user who does not have expert knowledge on the database storing method is provided with the workflow definition unit and the SQL definition unit, System.

In addition, by providing the data extraction module and the standard conversion module, it is possible to extract a standard format file for process analysis directly from the database without generating an intermediate level file. By providing the definition creation module, the workflow definition part and the SQL definition part There is an effect of providing a data preprocessing system capable of creating / modifying / deleting data, confirming / changing an SQL statement corresponding to a workflow definition subscheme, and selecting / excluding a specific scheme.

FIG. 1 is a diagram showing an MXML structure that is a standard input file format of a process mining program.
FIG. 2 is an exemplary diagram showing all event types (event types) and connection relationships allowed in MXML.
FIG. 3 is a block diagram illustrating a data preprocessing system according to an embodiment of the present invention.
FIG. 4 is an exemplary diagram illustrating a workflow definition used in a data preprocessing system according to an embodiment of the present invention.
FIG. 5 and FIG. 6 illustrate SQL statements used in a data preprocessing system according to an embodiment of the present invention.
7 is a block diagram illustrating a data preprocessing system according to another embodiment of the present invention.
FIG. 8 is a diagram illustrating an exemplary user interface used in a data preprocessing system according to another embodiment of the present invention.

The present invention relates to a data preprocessing system capable of extracting data from a database with a single click and generating standard format data for process analysis.

Data in a standard format for process analysis is data in a format that can be used in a program capable of analyzing a process, such as a data mining solution, and MXML is generally a standard format as shown in Fig.

In this MXML (Mining XML) structure, there are several process instances data under a process, and there are several pieces of audit trail entry data under one process instance. do.

A process instance refers to a case. In a container process, an individual container, a vessel, a truck, and the like may correspond to a process instance.

The audit trail entry refers to an event. In the container process, QC unloading job creation and QC unloading job completion may be included in the audit trail entry.

Below the Audit Trail entry are the Workflow Model Elements, Event Types, Time Stamps, Originators, and Attribute Data.

The workflow model element can be subdivided into individual actions that exist in the process and can be subdivided into event units. In the container process, QC drop, YQ dropout, YQ dropout, and the like may correspond to the workflow model element.

The event type specifies an event in the workflow model element when the workflow model element is subdivided into event units. Container processes can represent event types, for example, by job creation (schedule), completion (complete), and so on.

As shown in FIG. 2, there is an association (paper: A Meta Model for Process Mining Data) between event types that can be defined in MXML. The main flow in FIG. 2 has a Schedule-assign-start-complete flow, and it is possible to analyze the process without defining all the flows (for example, all workflow model elements can be defined by only one kind of complete event ).

The timestamp is the time when the event occurred, the originator is the actor that triggered the event, and the attribute is the attribute information that exists on all data except the time stamp and originator. Attributes can exist in both audit trail entries and process instances. Attributes for the object being analyzed are attributes of the process instance. Attributes for the behavior are attributes of the audit trail.

The data preprocessing system 100 for analyzing a process according to the present invention includes a definition module 110 in which a workflow definition and an SQL definition are recorded, a data extraction module 120 for extracting data from a database, And a standard conversion module 130 for converting the standard data into a standard format for generating a standard format for process analysis by extracting necessary data from the database as shown in FIG.

In more detail, the data preprocessing system 100 according to the present invention needs to retrieve necessary data from an event log stored in a database in order to extract data in the standard format as described above.

To this end, the definition module 110 of the data preprocessing system according to the present invention includes a workflow definition unit 111 for defining what data to extract and an SQL definition unit 112 for defining how to query the data in the database have.

More specifically, the workflow definition defines all objects for process analysis and events associated with each object. To this end, a process scheme (ProcessScheme) for querying the target process instance and at least one Audit Trail Scheme for querying each event, an audit trail entry under the process scheme, is included.

In addition, each scheme includes an ID (SqlMapperID) for searching for an SQL statement related to the schema, and an SQL statement corresponding to each ID is included in the SQL definition unit 112, so that data related to each scheme can be retrieved from the database .

Particularly, in the workflow definition unit according to the present invention, a time stamp (see FIG. 1) among data (see FIG. 1) that should be included in an audit trail entry such as a work model element, an event type, a time stamp, an originator, , And the originator are data having different values for each individual record. However, the work model element (Work Model Element) is used in that the event corresponds to an event in the process, , The event type is defined in the workflow definition scheme, thereby eliminating the need to include the workflow model element and event type for each individual SQL.

The SQL statements included in the workflow definition and the SQL definition section are illustrated in FIGS. 4 to 6 by way of example, and the workflow definition and the SQL definition can be written in the xml format as illustrated in the figure. Referring to these drawings, a detailed description will be given below.

FIG. 4 shows a process scheme for inquiring containers processed by the process from importing to exporting, and an audit trail scheme for inquiring an event related to the container (import device job creation, completion of the import device job) under the process scheme Respectively.

The process scheme of FIG. 4 includes a selectContainer which is an ID for matching an SQL statement for querying the containers processed by the process in the database, and the corresponding SQL statement as shown in FIG. ).

Also included in the audit trail scheme is an ID for matching an SQL statement for querying the event. An exemplary SQL statement corresponding to selectCompleteGateIn is shown in FIG.

In addition, as shown in FIG. 4, a workflow model element and an event type are defined in the audit trail scheme.

Since the workflow model element and the event type are included in the workflow definition scheme, data inquired by each SQL statement and SQL statement is included in the workflow model element and information about the event type It is possible to eliminate the unnecessary necessity of including it.

Referring to FIG. 5 showing a SQL definition for retrieving container data, a SQL definition for querying a process instance will be described. In an SQL statement for querying a target process instance, a variable (CASE_ID) It must be included. The conditional statement of the SQL statement must include the data query time range.

The reason why such information should be included is that the data of the object (container in FIG. 6) within the data inquiry time range can be retrieved by using the corresponding SQL statement, It is because it can distinguish.

Referring to FIG. 6 showing the SQL definition for retrieving the container import device completion event, the SQL definition for querying the audit trail entry will be described. In the SQL statement for querying the event process instance, Variable and event occurrence time (time stamp) must be included, and the SQL statement condition must include the data query time range.

The reason why this information should be included is that it is possible to inquire data of an event within the data inquiry time range by using the corresponding SQL statement, It is because we can distinguish what occurred at what time.

The data retrieval time range is applied to all SQL statements when a user enters a time range for retrieving data, so that objects and events within the retrieval time range can be retrieved.

Also, in FIG. 6, the time stamp is described as ATE_TIME, not Timestamp. This is because the Timestamp is used as a special meaning in the SQL statement of a specific database management system (DBMS). Therefore, the variable name of ATE_TIME is applied instead of this, and the variable name of CASE_ID described above is also changed . ≪ / RTI >

The data preprocessing system according to an embodiment of the present invention may use the data of the definition module 110 described above to input data from the database 200 connected to the data extraction module 120 .

To do this, the data extraction module must contain the database connection information and be able to extract data from the database using the SQL statements of the definition module. This data extraction module can be implemented using MyBatis plugin (Mybatis Plug_In).

In the data preprocessing system according to an embodiment of the present invention, ProM Import may be used as the standard conversion module 130. [ The program is an open source program developed by Eindhoven University of Technology which developed MXML format. It outputs the results obtained from SQL query in database to string file such as txt, csv, and provides this file as input of pro- You can get the file.

However, in the data preprocessing system according to an embodiment of the present invention, the data extraction module 120 is connected in the form of a plug-in before the standard conversion module 130, so that it is possible to perform the process analysis without the need to generate intermediate files such as txt, csv, You can get MXML files in standard format right away.

The data preprocessing system according to an embodiment of the present invention has been described with reference to the drawings. If you use this data preprocessing system, even if you are a non-expert who can not understand the structure of the database once the workflow definition and SQL definition are written by experts in advance, you can input the data for process analysis into standard Format.

The data preprocessing system according to another embodiment of the present invention may further include a definition creation module 140 for providing a user interface (UI) so that the user can modify the information stored in the definition module as shown in FIG. 7 have.

The definition creation module 140 provides a function to allow a user to create / modify / delete a workflow and an SQL statement stored in the definition module, which is connected to the definition module 110.

The definition module of the present invention not only provides a function that allows a user to create / modify / delete a process schedule and an audit trail scheme of a workflow, displays an SQL statement corresponding to the scheme, It provides the function to be able to modify it in case.

Also, it provides a function to select or exclude a specific scheme when extracting data, so that only data related to a specific object or a specific event can be extracted, and only relevant processes can be analyzed.

An exemplary form of the user interface provided by the above definition creation module 140 is shown in FIG. 8, and the user can use this user interface to create a process scheme for the analysis object and a scheme for the event Trail Scheme), adding / modifying / deleting schemes, displaying various information about selected schemes (including SQL statements), modifying, testing, and applying them.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but may be modified or altered by those skilled in the art without departing from the gist of the invention. Modifications and variations should also be regarded as falling within the scope of the following claims.

100: Data preprocessing system 110: Definition module
111: workflow definition unit 112: SQL definition unit
120: data extraction module 130: standard conversion module
140: definition creation module 200: database

Claims (9)

A definition module including a workflow definition unit that defines what data to extract and an SQL definition unit that defines how to query the data in the database;
A data extraction module for extracting data from a database using information recorded in the definition module; And
A standard conversion module that receives data from the data extraction module and converts the data into a standard format file for process analysis;
, ≪ / RTI >
The workflow definition unit includes a schema for querying an object related to the process and a schema for querying an event related to the object. Each scheme includes an identifier (ID) for searching an SQL statement corresponding to each scheme from the SQL definition unit, ), And a workflow model element and an event type are defined in a scheme for inquiring the event.
delete delete The method according to claim 1,
Wherein the SQL statement includes a variable for querying the ID of the target and includes a data query time range in the conditional statement.
The method according to claim 1,
Wherein the SQL statement for querying an event includes a variable for inquiring an ID of a related object and an event occurrence time, and includes a data query time range in a conditional statement.
The method according to claim 4 or 5,
Wherein the data inquiry time range is the same for all SQL statements when a desired data inquiry time range is input.
The method according to claim 1,
Further comprising a definition creation module for providing a user interface that allows a user to create, modify, and delete data stored in the workflow definition unit or the SQL definition unit.
8. The method of claim 7,
Wherein the user interface provides SQL definition display, modification and testing functions corresponding to the schemes.
8. The method of claim 7,
Wherein the user interface provides a function to select or exclude a specific scheme in extracting data.

Images