DocXtractor II English
-
Upload
ichsanul-anam -
Category
Documents
-
view
52 -
download
2
Transcript of DocXtractor II English
© ELO Digital Office GmbH
DocXtractor INVOICE
Automated incoming mail processingand business process optimisation
© ELO Digital Office GmbH
© ELO Digital Office GmbH
structured andunstructured informationof any source
structured andunstructured informationof any source
making documents work ... making documents work ...
to capture informationto capture information
to provide informationto provide information
to organise informationto organise information
© ELO Digital Office GmbH
ELO Digital Office – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
Questions
2
3
4
5
ELO Digital Office – About us1
6
© ELO Digital Office GmbH
Business objectives: Intelligent document processing and business process optimisation
Market entry: 1995
Main product: ELO ECM-Suite, DocXtractor
Target market: Insurance, banking,retail, manufacturing
Subsidaries: Stuttgart (head office) Hamburg, Dortmund, Munich, Gera
Luxemburg, Belgien, Nederlands, France, Poland, Tschech, Italy, Australia, Hungary, Austria, Turky Switzerland,…..
ELO Digital Office GmbH is a market leader for EnterpriseContentMangement software and input management
© ELO Digital Office GmbH
History
History of ELO Digital Office GmbH
© ELO Digital Office GmbH
Competence and experience – ELO Digital Office ….
… is an expert for intelligent document processing and business process optimisation.
… offers content capturing software for complex applications.
… solutions allow the automatic categorisation and analysis of any structured and unstructured documents and the qualified extraction of information contained.
… achieves highest rates of recognition and data quality by the implementation of innovative technologies.
… solutions reduce costs in document processing areas of a company.
© ELO Digital Office GmbH
© ELO Digital Office GmbH
ELO Digital Office – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
Questions
1
3
4
5
Incoming mail processing – The challenge2
6
© ELO Digital Office GmbH
Objectives of Document Related Technologies
Meaning of Document Related
Technologies:
Capturing as needed, systematical organisation as well as appropriate access to all information
Connection between document and business process
Resulting business objectives:
Efficient information processing
Safety of quality and efficiency of processes, decisions, products etc.
Prevention of misallocation of resources
Creation and assurance of competitive advantages
to captureinformation
to organiseinformation
to provideinformation
© ELO Digital Office GmbH
input media: paper, email, fax etc.
forms semi structured documents free-form documents
© ELO Digital Office GmbH
Connection between data, information and knowledge
image
image objects layout structure
characters
„d“ „S“ „2“data
interpretationINFORMATION
informationpresentation
sender recipient date subject signature ...
sender recipient date subject signature ...
logical objects
offer offer
message type
order order
invoice invoice
... ...
business data
processeswords
knowledge
© ELO Digital Office GmbH
dat
a C
aptu
re v
iew
b
usi
nes
s p
roce
ss v
iew
business processes
Company
Customer
data capture
?
forms & free structured documents
paper
fax
etc.
field 2
field 1
field 3
Challenge of free form incoming mail processing
Heterogeneous documents
High daily volume
Growing amount of free form documents
Documents are central input factor of business processes
Today business process design is often isolated from document processing steps
© ELO Digital Office GmbH
Technical success factors for document processing
Processing documents efficiently +
Process qualityProcess quality
Process flexibilityProcess flexibility
+
+
Process speedProcess speed
Process acceleration and automation
Adherence to time and dates
Cash flow optimisation
. . .
Process acceleration and automation
Adherence to time and dates
Cash flow optimisation
. . .
High data quality with minimum verification efforts
Homogenous data (consistency with ERP/ host systems)
Document processing as an end-to-end solution
High data quality with minimum verification efforts
Homogenous data (consistency with ERP/ host systems)
Document processing as an end-to-end solution
Processing of all documents from any customer
No customisation needed for new customers
Flexibility with regard to data extraction
. . .
Processing of all documents from any customer
No customisation needed for new customers
Flexibility with regard to data extraction
. . .
© ELO Digital Office GmbH
Economic sucess factors for document processing
Short payback period of investment +
Acquisition costsAcquisition costs
Process costsProcess costs
+
+
Investment costsInvestment costs
Software costs
Hardware costs
Project planning costs (internal und external)
. . .
Software costs
Hardware costs
Project planning costs (internal und external)
. . .
Initial installation costs
Verification costs
Data quality
. . .
Initial installation costs
Verification costs
Data quality
. . .
Customer specific adaption effort
Servicing costs
Flexibility regarding fields extracted
. . .
Customer specific adaption effort
Servicing costs
Flexibility regarding fields extracted
. . .
© ELO Digital Office GmbH
processing with human interaction
processing without human interaction
recognition
extraction
scanning
indexing
automatic data
transfer
manual process-
ing
automaticdata transfer
manual processing
manual processing
manual distribution
electronicaldistribution
incoming mail recognition distribution processing outgoing mail
text and print system
archive / document management system (archive / DMS)
process management tool
CRM system
electronical documents
telephone
paper
fax
internet
© ELO Digital Office GmbH
ELO Digital Office GmbH Technologies – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
Questions
1
2
4
5
Business process optimisation – The solution3
6
© ELO Digital Office GmbH
OCR-ICR
company
OCR-ICR
company
Support of automatic business transaction
classification
extraction
approach
customer
A Customer sends documents unrequested e.g. notice of loss
common expectations
expectation
B Customised business transaction already exists e.g. confirmation of address change
company
call center
classification
extraction
plausibility
specific expectationsp1
p2
p3
customer
expectation
© ELO Digital Office GmbH
OCR / ICR system as an integrated component for business process optimisation
index
info
Kai KornBergstr. 2467659 KL
cancellationaccident insur.
new address
process
cancellation
accident insur.
change of address
insurance holder
key data
police number :
1258 KK 12 U 8
address new
1258 KK 1154
Proccessing of heterogenous incoming mail
Short processing time > High customer satisfaction
Efficient business process organisation and optimisation
© ELO Digital Office GmbH
Selected requirements of intelligent OCR / ICR solutions
Controlling of the whole document processing from scanning to data storage with high system stability
Processing of heterogenous batches of documents and also electronic documents
Automatic designation of classification features for free-form documents
Extraction of customer information depending on business process
Quality increase of captured information by mathematical and logical checks
Integration of business databases for validation purposes
Support of automated processing without human interaction
High scalability by outsourcing load intensive processes to external clients
Minimal effort for adaptation of new document classes
© ELO Digital Office GmbH
ELO Digital Office GmbH Technologies – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
Questions
1
2
3
DocXtractor – The product4
6
5
© ELO Digital Office GmbH
Highlights of incoming mail processing with DocXtractor
Processing of whole heterogeneous incoming mail (paper, fax, email, electronic documents) without any explicit presorting
Minimal training and implementation effort – complete GUI-based training and testing
Minimal administration effort – administration and monitoring completely in customers hands
Self-learning and self optimizing system with auto-adaptive, intuitiv and visual administration and configuration support
Substantial statistical analysis and reporting in test environment as well as in production for performance measurement and ressource planning
© ELO Digital Office GmbH
Workflow
Document process with DocXtractor (internal workflow)
automated document processing with DocXtractor
workflow
database
ERP
archive
automatedprocessing
or agent
verification workplace
training workplace
automated
analysis
fax server
email server
scanner exportpaper
fax
electronic documents
export import
DocXtractor automates the classification process and provides the required information automatically.
import
administration workplace
© ELO Digital Office GmbH
Workflow
Document process with DocXtractor and ELOscan (internal workflow)
automated document processing with DocXtractor
workflow
database
ERP
archive
automatedprocessing
or agent
verification workplace
training workplace
automated
analysis
export
export import
DocXtractor automates the classification process and provides the required information automatically.
import
administration workplace
fax server
email server
paper
fax
electronic documents
SCAN import
© ELO Digital Office GmbH
DocXtractor
Image preprocessing
DocXtractor prepares image files for an optimal recognition
© ELO Digital Office GmbH
Classification can be performed using different methods (AutoClassifier, layout, search patterns, tables, ...)
commercial invoice
medical invoice
insurance contracts
bank account changes
etc.
address changes
Using AutoClassifier the classification criteria can be generated automatically during the training process.
© ELO Digital Office GmbH
OCR result field name
Data extraction
localisation of data fields
7929418
P e tz, Erwin
94,80
190,80
8,16
44,0 8
337,82
invoice number
name
position 1
position 2
amount for disposal
VAT
total amount
Information extraction based on forms
© ELO Digital Office GmbH
ELO Digital Office GmbH Top Down Search
company name street z.code city bank code account
Thomas Cook AG Zimmerstr 61440 Oberursel 20041111 4786543
Adolf Würth GmbH Postfach 74650Künzelsau62091800 10681000
Voith AG Pöltenerstr 74650 Künzelsau62091800 21389700
BMW AG Pacalstr. 70569 Stuttgart 70540660 518378908
.....
master data
Knowledge about location of fields is not necessary
Perfect fit for free form documents and invoices
Fuzzy search (tolerant against OCR errors and different spellings)
Optimal results without training
© ELO Digital Office GmbH
Automated quality assurance and validation of information
logical
checks
logical
checks
mathematical checksmathematical checks
7929418
P ee tz, Erwin corr.
94,80
190,80
8,16
44,0 6 corr.
337,82
matching with
master data
matching with
master data
7929418
P e tz, Erwin
94,80
190,80
8,16
44,0 8
337,82
invoice number
name
position 1
position 2
amount for disposal
VAT
total amount
© ELO Digital Office GmbH
Manual verification of information
7929418
logical checkslogical checks
matching with master data
matching with master data
mathematical checksmathematical checks
Peetz, Erwin
94,80
190,80
8,16
44,06
337,82
invoice number
name
position 1
position 2
amount for disposal
VAT
total amount
quality assured
data export
Manual data verification will also use automatic validation processes to improve data quality
© ELO Digital Office GmbH
USPs of DocXtractor
DocXtractor is a product for automated processing of the whole incoming mail
Standardised interfaces to archive-, DMS-, ERP- and workflow systems as well as capturing solutions simplify the integration
Customer oriented and continuous development with fixed release dates
Cooperation with customers in Product User Groups
Extensive service offer
Customising by system configuration (coding and compilation are not necessary)
Release independent integration of technical requirements is possible
Market-leading methods of classification and extraction for reduction of manual effort of verification
© ELO Digital Office GmbH
Capability characteristics
Capability characteristics of DocXtractor
Controlling of whole document handling from scanning to data storage with high failure safety
Processing without manual presorting of documents
Processing of electronic documents
Automated definition of classification characteristics for free form documents
Extraction of customer information dependent on business process
Quality improvement of selected information by mathematical and logical checks
Integration of business database for output validation
Support of automated process control (processing without human interaction)
High scalability on a client-server-architecture
© ELO Digital Office GmbH
ELO Digital Office GmbH Technologies – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
Questions
1
2
3
DocXtractor – The system architecture
4
64
54
© ELO Digital Office GmbH
DocXtractor SUITE : Components and modules
Legende
Administration / Konfiguration
Q-Sicherung/Statistik
Testsysteme
Document Manager
Import Analyse
FREE FORM
ExportAdaptionen Verifikation
SAP-Module
MonitoringImport/Export
Archiv/DMSECM
E-Doc/Exchange
File/ScanningXML
BASIC
INVOICE
ORDER
PKV
Verifier
Supervisor
Archiv/DMSECM
Datenbank
File/XML
Document Finder
Reporting
Archiving
•Modul 1
•Modul ...
•Modul 2
Components
•Modul 1Module 1
Module ...
•Modul 2Module 2
Legend
Administration / Configuration
Quality security / Statistic
Test systems
Document Manager
Import Analysis
FREE FORM
ExportAdaption Verification
SAP modules
MonitoringImport/Export
Archive/DMSECM
E-Doc/Exchange
File/ScanningXML
BASIC
INVOICE
ORDER
PKV
Verifier
Supervisor
Archive/DMSECM
Database
File/XML
Document Finder
Reporting
Archiving
© ELO Digital Office GmbH
Internal system workflow DocXtractor
Analyser
image preprocessingimage preprocessing
Document
Manager
Document
Manager
classification classification
information extraction information extraction
validation and correctionvalidation and correction
customer DB customer DB
Coordinator
ext. application ext. application
Exporter Exporter
image source image source
Verifier
SupervisorSupervisor
OCRcorrection
OCRcorrection
document
definitions
document
definitions
DocXtractor DB matching DBmatching DB result DBresult DBcontrol DBcontrol DB
Importer ImporterThe database oriented
document analysis
ensures a consistent
system
© ELO Digital Office GmbH
The client server architecture guarantees a high failure safety in conjunction with the necessary scalability
DocXtractor server
Analyser 1
.
.
.
.
.
.
.
.
.
.
Coordinator
Importer
DocXtractor DB
Analyser m
Verifier 1
Verifier m
Exporter
© ELO Digital Office GmbH
Client ability
DocXtractor supports the process of different clients in one system
Every sub system can have its own workflow and its individual configuration
DocXtractor
incoming mail
client 1
customer system
client 1
sub system f
client 1
incoming mail
client n client n
customer system sub system f
client n
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
© ELO Digital Office GmbH
Technical requirements
Technical requirements
Server Processor: 2* Pentium IV 3,0 GHz or higher, poss. DUAL Core RAM: min. 1 GB per processor Hard disk: min. 2*30 GB, mirrored and failsafe Operating system: Windows 2000 Server (Advanced), Windows 2003 Server
(Enterprise) Software: MS SQL Server, Oracle 9, 10 (Server), IBM Informix (Server), IBM DB2
(Server), or external
Analysis Clients Processor: 2* Pentium IV 3,0 GHz or higher RAM: min. 1 GB per processor, hard disk: min. 10 GB Operating system (alternative): Windows 2000 Professional, Windows 2003 Server (Enterprise) Software: MS SQL (ODBC), Oracle 9, 10 (ODBC), IBM Informix (ODBC), IBM DB2 (ODBC)
Document Manager Client / Verifier Clients Processor: 1* Pentium IV 2,4 GHz or higher RAM: min. 1 GB, hard disk: min. 10 GB Operating system: Windows 2000 Professional, Windows XP Software: MS SQL (ODBC), Oracle 9,10 (ODBC), IBM Informix (ODBC), IBM DB2 (ODBC)
Other equipment Network hard disk (100 Mbit/s)
© ELO Digital Office GmbH
ELO Digital Office GmbH Technologies – About us
Incoming mail processing – The challenge
Business process optimisation – The solution
DocXtractor – The product
DocXtractor – The system architecture
QuestionsQuestions
1
2
3
4
64
54
© ELO Digital Office GmbH
Thank you for your attention