1 IBM Datacap Taskmaster Capture Tom Simalchik, Capture Offering Manager.
-
Upload
moses-ellis -
Category
Documents
-
view
273 -
download
10
Transcript of 1 IBM Datacap Taskmaster Capture Tom Simalchik, Capture Offering Manager.
1
IBM Datacap Taskmaster Capture
Tom Simalchik, Capture Offering Manager
Disclaimer
© Copyright IBM Corporation 2011. All rights reserved.U.S. Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
THE INFORMATION CONTAINED IN THIS PRESENTATION IS PROVIDED FOR INFORMATIONAL PURPOSES ONLY. WHILE EFFORTS WERE MADE TO VERIFY THE COMPLETENESS AND ACCURACY OF THE INFORMATION CONTAINED IN THIS PRESENTATION, IT IS PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED. IN ADDITION, THIS INFORMATION IS BASED ON IBM’S CURRENT PRODUCT PLANS AND STRATEGY, WHICH ARE SUBJECT TO CHANGE BY IBM WITHOUT NOTICE. IBM SHALL NOT BE RESPONSIBLE FOR ANY DAMAGES ARISING OUT OF THE USE OF, OR OTHERWISE RELATED TO, THIS PRESENTATION OR ANY OTHER DOCUMENTATION. NOTHING CONTAINED IN THIS PRESENTATION IS INTENDED TO, NOR SHALL HAVE THE EFFECT OF, CREATING ANY WARRANTIES OR REPRESENTATIONS FROM IBM (OR ITS SUPPLIERS OR LICENSORS), OR ALTERING THE TERMS AND CONDITIONS OF ANY AGREEMENT OR LICENSE GOVERNING THE USE OF IBM PRODUCTS AND/OR SOFTWARE.
IBM, the IBM logo, ibm.com, FileNet, Datacap and IBM FileNet Capture, Taskmaster, Rulerunner and FastDoc Capture are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml
Microsoft SharePoint, EMC, Open Text, Oracle, IBML, AIIM, Kinetic, Computerworld and Smithsonian are trademarks or registered trademarks of their respective companies or organizations. Other company, product, or service names may be trademarks or service marks of others.
3
Agenda• The Importance of Document Capture
• IBM Datacap Taskmaster Capture Update
• Customer Case Studies
A Transformation is Happening in ECM
DefensibleDefensible
AccessibleAccessible
Competitive Competitive AdvantageAdvantage
CollaborativCollaborativee
RelevantRelevant
InsightfulInsightful
ContextualContextualIT
LegalRecords
Information Management
(RIM)
Line of Business
…To Systems of Engagement.
From Systems of Record….
5
Capture it.
Analyze it.Activate it.
Socialize it.Govern it.
Organizations who put Content In Motion Can Take Advantage Of the Full
Spectrum of ECM Solutions
High Value solutions spanning multiple industries
• Advanced case mgmt
• Customer Service / Experience Mgmt
• Account Opening & Management
• Courts and Justice
• Claims Processing & Optimization
• Benefits Adjudication
• Insurance Underwriting
• Loan Origination / Mortgage Processing
• Social content mgmt
• Human Capital Management
• Education Intervention Management
• Content Search and Analytics
• Voice of the Customer
• Patient Diagnostics & Care Coordination
• Government and Crime Intelligence
• Enterprise Fraud Management
• Defensible Disposal & Value Based Archiving
• Retention & Records Management
• eDiscovery
Content at Rest = Cost, Content in Motion = Value
CAPTURECAPTURE SOCIALIZESOCIALIZE GOVERNGOVERNACTIVATEACTIVATE ANALYZEANALYZE• Document Imaging
and Intelligent Document Capture
• Enterprise Platform Services
• Enterprise Report Management
• Document Classification
• Accounts Payable
• Medical Claims Processing
• Distributed scanning
IBM ECM Foundational Solutions for IT. Compliance & Legal Buyers
IBM ECM Industry Specific Solutions targeting LOB and New Buyers
IBM ECM Cross-Industry Solutions targeting LOB & New Buyers
7
Capture is the Critical Onramp for Content
• Better customer/vendor service and communications
• Reduced time and resources required to manage paper and related business processes
• Improved cash flow, reduced transaction and paper costs while growing the business
• Improved collaboration as documents can be immediately accessed and shared around the world
• Elimination of lost files• Secure and reliable backup and disaster
recovery
• And overall Return On Investment for Systems of Engagement
How do Customers Achieve their ROI goals?
• Reduce cost of transporting paper to a central location
– Scan documents in remote locations – branches, stores, offices, etc.
– Savings can be more than $1M annually
– Key capability - Distributed Capture
• Reduce data entry labor costs
– Extract data from documents without manual keying
– Potential to reduce data entry staff up to 90%
– Large organizations can have hundreds of employees performing data entry
– Key capabilities – Rules, Advanced Data Extraction
• Reduce cost of document capture
– Reduce paper sorting and document preparation
– Potential to reduce capture staff up to 50%
– Key capabilities – Rules, Advanced Data Extraction
• Standardize
– Single vendor ECM and Capture solution
– Replace obsolete or costly legacy capture systems
– Reduce license fees, support and maintenance costs
– Eliminate volume-based pricing
9
Components of Enterprise Capture
Copyright 2009 Harvey Spencer Associates, Inc
Field
Field
Branch
CentralMallroom Department
Fax eMail
10
Strategic Nature of Capture
• Capture applications are the gateway to enterprise content strategies
• Driven by several key value components:– FTE reduction / repurposing– Data entry error reduction– Document transportation costs– Document retention costs
• Growing document production (paper and electronic) and government regulation mean that Capture/ECM projects remain viable and justifiable even in uncertain economic times
11
Datacap Taskmaster Capture Update
12
IBM Vision of Enterprise Capture
A universal capture portal that can transform all documents
Capture documents at every entry point in the Enterprise
Input any mode for consistent processing rules
Point and click capture process management enables clients to orchestrate complex capture solutions – without waiting for expensive programmers to build an application
IBM Datacap Taskmaster V8.01
• Automatic document recognition, classification and data extraction
• Web support for distributed deployments
• Optimized manual data entry• Flexible functional security• Data lookup capability• Powerful background processing• SOA via Web Services• Feeds line of business systems and ERP• Advanced Account Payable
Advanced Document & Data Capture
IBM Datacap Taskmaster V8.01
• Export to IBM FileNet P8, IS and CM8• Support for non-IBM repositories from
EMC, Open Text, Oracle, Microsoft and others with generic file/XML
• Scanned documents as well as electronic documents
Advanced Document & Data Capture
Capture ProcessScan or Import documents.
Classification - enhance & identify each individual page
Organize the individual page into documents
Extract barcodes, machine print and hand printed data
Validate and supplement data using rules and database lookups
Verify documents with exceptions
Export data to business systems and documents to ECM systems
Page Input• Scan paper documents operating scanners directly
– Thick and thin client scan user interfaces– Uses standard drivers: TWAIN, ISIS
• Import / Vscan– Interactive thick and thin client import user interface– Unattended continuous import on background server processes – Sources
• file system• fax connector to Rightfax ***• email connector to IMAP and Exchange ***
• Format conversions– Converts files to single page TIFF format for internal processing– Retains original input files – Converts images
• Color, gray scale, and bitonal TIFF, JPEG, PDF, PNG
– Converts electronic documents***• MS Word, MS Excel, MS Outlook Message & Zip
*** separately charged components
Page Identification• Classifies pages using multiple methods
– Structure – known or expected page ordering– Barcode matching– Image pattern match e.g. logos, anchors– Fingerprint matching – image or text– Text search for regular expressions or key phrases– Text analytics using IBM Classification Module connector ***– OCR can be done on-the-fly or skipped
• Enhances Images– Deskew – Despeckle. remove noise, lines, smears, and borders– Enhance characters
• Pre-processing Options– Crop out portions of images– Split single images into multiple images
*** separately charged components
Page Identification: Smart Separator Sheet
• Document / Form type barcode
• Additional Data– Could be pre-printed– Or entered by user
Page Identification: Pattern Recognition
• Very fast matching to unique marks on a page – “anchors”• Used with fixed forms• Most commonly used with ICR – handprint forms• PatternMatchIdentify Action
Page Identification: Fingerprint Recognition
• Fast (sub-second) – does not require OCR• Matches the patterns of light and dark - Characters, blobs, words, text
lines• Supports thousands of stored page templates• Also differentiates between multiple formats of the same page type• Adjusts the positions of zoned fields• FindFingerprint Action
• Scanned Image FingerprintComparing patterns of light and
dark
Page Identification: Keyword
• Following OCR to recognize machine print text on a page• Regular expressions find key words and phrases• Search zones or search the entire page• Searches can be stored externally in key files
\bSettlement\s*Statement.*HUD.*[1]\b
Page Identification: Connector to Classification Module
• Taskmaster – Extracts text using OCR – Optical Character Recognition– Calls Classification Module to identify the page
• Classification Module analyzes the text content– Uses natural language processing and semantic analysis– Assigns confidence score to each category suggestion (0 – 100)– Returns the classification results to Taskmaster
Page Identification: How does Classification Work
• Taskmaster examines each page using multiple methods– The fastest methods are done first : barcode, pattern match, & fingerprint– The slower methods that require OCR follow: Text analytics and keywords– Finally rules examine the context to determine if any remaining pages can be
identified based on the surrounding pages
• The Taskmaster document hierarchy specifies page types contained in each document– Separates and assembles the pages into documents
• The system outputs classification results statistics to support optimization • Feedback loop improves future results
– Image fingerprints populated to fingerprint database– Text classification trained with feedback to analytics engine
• Exceptions, low confidence results are reviewed and classified by users
Document Assembly• Create logical documents that consist of one or more pages.
– The system groups the pages into documents and can checks if the resulting structure is valid
• Separate documents using– Page Identification / classification– Barcodes / patch codes– Rules
Data Recognition• Character Recognition (OCR/ICR)
– 3 Recognition engines included in base product– Machine print and hand print– Zonal fields– Regular expression text search– Full page text– Learns field locations from the end-user interaction– Dual engine voting
• Handwriting Recognition***– Cursive & hand print– Word recognition reads whole words or phrases. – Improves recognition by using application-specific context
• Optical Mark Recognition (OMR)– Check boxes, bubbles, or the presence of a signature
• Bar Code recognition – 1D: 2 of 5, Interleaved 2 of 5, Airline 2 of 5, Matrix, Matrix 2 of 5, Code 32,
Code 39, Code 39 Extended, Codabar, Code 93, Code 93 Extended, Code 128, EAN13, EAN8, UPC-A, UPC-E, Addon 5, Addon 2, UCC128/EAN128, Patch Code, PostNet
– 2D: PDF417, Datamatrix, QR
*** additional license required
Data Validation• Checks accuracy and flags errors• Validation can include
– Self checking mechanisms such as field patterns, field lengths, formats, and check digits
– Valid ranges, choice lists, and checking calculated values– Validating field values against business rules– Database lookups– Confidence thresholds
• Languages Supported: Portuguese (Brazilian), French, Spanish (Castilian), German, Italian, Swedish, Dutch, Polish, Czech, Slovak, Romanian, Croatian, Hungarian and Turkish
Data Verification• Display exceptions for review and correction by human operators• User Interfaces
– Windows thick client– Taskmaster Web – through Internet Explorer web browser
• Key capabilities:– Click ‘n Key – select and fill-in data by clicking and selecting on the
image display– Learns where data was found automates the next time– Optionally display only pages with exceptions– Image snippets and color coded confidence levels– Multi-pass & blind verification– Line item details– Keyboard shortcuts for high-speed keying without the mouse– Image rescan
Verification User Interface Screen
High-Density Screens and Click N’ Key
Data Export• Export Documents
– IBM FileNet CM, IBM FileNet Image Services, IBM Content Manager– EMC Documentum***, OpenText LiveLink***, Microsoft Sharepoint *** – others via file system export or custom actions
• Export Data– XML and text files– Database updates– Use web services via custom actions (requires customization)
• Formats– TIFF, JPEG, PDF (image-only, or w/ searchable text), PDF/A
• Original input files and unenhanced images are retained and can be exported
*** separately charged components
Datacap Taskmaster Accounts Payable Capture V8.0.1
• Preconfigured application
• Captures, verifies and routes without manual data entry
• Locate and extract data including header and line item detail
• Learns new invoice types from operator
• Accurately captures all line items, even multi-page
• Complex validation rules on dates, math, lookups, data types, etc.
• Look up vendors, add line items, locate line items, calculate missing values
• Aids three-way match with Purchase Order Line item Reconciliation
• Send to operator for handling exceptions
Taskmaster Accounts Payable Capture Advantages
• No preproduction set-up required
• Adapts to new invoice layouts on-the-fly learning the first time
• Single page, multi-page, attachments
• Line item capture out of the box
• POLR – Purchase Order Line Item Reconciliation - streamlines 3 way match downstream
• Thick and thin client architecture and user interfaces
• Fingerprint Service accommodates tens of thousands of vendors
• Pricing model by user - NOT pages/documents scanned or processed
• ROI in 6 – 12 months
• Many years experience in AP automation
• Easily extensible to new document types and add-on applications, i.e. sales orders, remittances, etc.
Datacap Taskmaster Medical Claim Capture V8.0.1
• Capture CMS 1500 medical claims and UB-04 institutional claims– Preconfigured capture for 100% of fields on the CMS 1500 (aka “Professional”)– Complete capture of all fields on the UB-04 claim (aka “Institutional”)– Plus attachments
• Thin Web and thick Windows clients• Support for black claims• Validations
– Lookups – i.e. Match diagnosis and CPT codes– Business rules– Math calculations– HIPAA compliant 837 EDI output
• Browser-based scanning, verification and application administration and reporting
• Extendible to other claim types and beyond claims to other documents
Benefits: Improve Accuracy and Efficiency
• Document automation can double data entry productivity!• OCR increases data accuracy• Data entry cost can be reduced by 50% and more
– Human operator = 200-240 claims/day*– IBM Datacap Taskmaster = 600+ claims/day
• Rapid deployment delivers faster ROI• Reduced processing time provides live data to the enterprise faster for
better visibility• Improved customer service from image enablement
– Majority of claim inquiries can be answered during initial call
• *Source Health Data Management
IBM Datacap Taskmaster Enterprise Expansion Options
• New: Advanced text classification with IBM Classification Module• IBM Datacap Rulerunner Enterprise – enterprise scalability through
virtualization• Connectors for eMail and Electronic Documents
– Access mail server(s) via Internet Message Access Protocol (IMAP), which is supported by IBM Lotus Domino, Microsoft Exchange Server, Novell GroupWise, and other mail servers
– Provides ability to convert Microsoft Word, Excel, Outlook, PDF, and multipage TIFF files to single page TIFF files for capture processing
– Supports extraction of ZIP archives
• Connector for Fax
• Connectors for non-IBM repositories: EMC Documentum, Microsoft SharePoint and OpenText LiveLink
36
Datacap Taskmaster Capture
Customer Case Studies
37
Murphy-Hoffman Trucking Company Eliminates Shipping Costs
• 65 regional sales and service centers throughout the Midwest
• Replaced overnight shipping expense with 65 scanners and IBM Datacap Taskmaster Capture for browser-based scanning
• Invoices, sales, lease and service documents are scanned as soon as they are generated
• Uploaded to Kansas City headquarters for processing and storage
• Now documents are available immediately
• Staff at headquarters no longer wait for documents to arrive
• Document shipping expense eliminated
37
“Now staff at headquarters isn’t waiting until paper arrives to perform their work. They always have work available.” – – Imaging Manager, Midwestern Trucking CompanyImaging Manager, Midwestern Trucking Company
38
Virginia Department of Taxation enables workers in low income areas“We realized we could use the thin client to have at-home workers do data entry and verification of returns.”
— Nancy Wilson, Virginia Tax’s Manager of Automated Nancy Wilson, Virginia Tax’s Manager of Automated Processing SystemsProcessing Systems
• Processing 1.5 million paper tax returns every year, scanned in Richmond processing center
• Each captured return is presented to a verify operator who confirms data accuracy and fixes low confidence characters when needed
• Virginia passed a law in 2008 to stimulate jobs in low income areas
• Virginia Tax distributes a percentage of tax returns to At-home workers in low income areas using Datacap Taskmaster Capture’s browser-based verify panel
• Distributed capture helps Virginia deliver on its economic pledge and provides Virginia tax processing staff with maximum flexibility
38
39
BlueCross BlueShield Health Insurer Captures All Documents
39
• Purchased IBM Datacap Taskmaster Medical Claims Capture to automate input of 12,000 paper health claims a day
• Reduced labor by 50% and shrunk turnaround time
• Made a strategic decision to extend capture to other departments
• Began a process of adding one or two departments a year
• Now they are scanning 50,000 documents a day, including:
• Contracts
• Enrollments
• Medical tests
• Invoices
• Human Resources
• Added remote scanning from 5 satellite offices
• Added fax capture
• Email and Electronic documents on horizon
“I can see us adding new documents to our capture portal for a very long time.”
— Claims Manager, Major BCBSClaims Manager, Major BCBS
40
Global Logistics Company improves productivity and service
150,000 documents arriving every day from every source – mail, fax, email - and piling up rapidly as company prepared customs paperwork for shipments. Customs has many requirements for complete declaration at border crossing
Deployed seven imaging applications enabling faster order processing with fewer errors
Process ~600,000 pages per day in U.S. (~3,000 users) and expect to process ~4 million pages per day (~10,000 users) globally.
Company is able to move more shipments across borders with 30% less resources with reduced lost documents and data errors while also improving cycle times and accuracy.
40
Represents the state of the art for capture today: capturing paper, fax and emails, distributed scanning from many different sites, with many rules-driven variations.
Any questions?
More info from:Tom Simalchik – [email protected] Twigg – [email protected]