Collaborate 2012-business data transformation and consolidation for a global energy services company

Business Data Transformation and Consolidation for aGlobal Energy Services Company

Presented By:John PerkinsDirector, ChainSys Corporation, Lansing MIwww.chainsys.com

IntroductionBusiness and organizationalchangeshave significantramificationson data requirementsfor a globalcompany.Combine these changes with radical business process change and you have a functionalandtechnicalproblem with increased complexity. Migrationof master, referentialand transactionaldata tonew business organizationsrequire precise planning,coordinationand execution. We will review in detaila case study for a global energy services companyreviewing the process followedthroughoutthe project.Objective 1: To provide project and data strategies to dealing with multiple organizationalconsolidation.Objective 2: To discuss both manual and automatedtools that were used to create the new businessentity.

Objective 3: To review the structured process and techniquesfollowedby the project team.Objective 4: To review lessons learned in dealingwith a complexbusiness process and dataconsolidationproject.

Objective 5: To review processes to followto mitigate business risks with data migration.

Intended Audiences:(i) Individual contributor (ii) Project team member (iii) Project Manager

Project Overview

Established Power Generation Company Part of USD 4Billion Group.Operational capacity of 191 MW of power and generating capacity of2460 MW under various stages of implementation. Demerged from BILT.Captive power plants operating since 1955.

PL

AN

T4

PL

AN

T2

PL

AN

T1

PL

AN

T3

Project Overview

APIL

CPP

HO THERMALPOWERPLANTS

IPP

Korba WestPower Co Ltd

Jhabua PowerLtd

LEGEND:APIL- Avantha Power and

Infrastructure LimitedCPP – Captive Power Plants

IPP - Independent Power PlantsHO – Head Office

Project Overview

Legacy System

Oracle EBS 11.5.7Modules:

Oracle OPMOracle Financials

New SystemOracle EBS R12.1.3

Modules:Oracle OPM

Oracle FinancialsOracle Projects

Oracle EAMOracle OM/PO/QA/QP

Why Reimplementation?

Demerger from BILT in 2006 and need for separate legal entity andfinancial reportingNeed to re-engineer the E2E processes & standardizationsRequirement to capture production and product costs for pricing to coverROI mark up and for regulatory reportingNeed to address requirementsrelating to capturing of Project Costs,Asset Capitalizations and Maintenance CostsNeed to remove the redundant workarounds by enabling the newfunctionalities

Challenges…

Operational

• Multiple financial transaction

• Business process re-engineering• No visibility into production cost• No tax benefits• No tracking of project expenditures and

capital expenditures• Complicated Period closing cycle• Valuation for Coal and Chemicals not

available.

Compliance and Governance

• Comply with IFRS

• Comply with Tax rules• Improve Group level

consolidation and financialreporting

• Comply with multiplelegislative, industry andgeography requirements.

Technology

• End of life application – OracleE-BS 11.5.7

• No good Reporting / BI tools• Inadequate MIS Reporting

Process

Improvements

Process

Improvements

Key Concepts

Plan Obtain Store Maintain Apply DisposeData Life Cycle

Master data

•Customers•Employees•Vendors•Locations•Organization

•Accounts•Products

Reference Data

•Customer type•Item type

Transactional Data

•Sales orders•Purchaseorders•Trips•Deliveries•Invoices•Payments

Meta Data

•Table name•Field name•Constraints•Data types

Data Mart

•Extracted data•Transformation•Validations•Cleansing•Data Quality

Data Manipulation

•Extraction•Validation•Transformation•Cleansing•Consolidation

•Cross Reference•Loading

Data Migration is Complete

5

Data is Validated and

Incorrect Data Rejected

Data is Extracted to the Data Mart

Data is Transformed and

CleansedData File Updated

Target System

Define Data Objects and Attributes to be Migrated,Data Validation, Transformation and Cleansing Rules and Requirements

appMIGRATEData Staging Table

Source System

appMIGRATE Templates

are Configured

appMIGRATEData Staging Table

Data

StagingTable

Data

Staging

Table

Error Exception API’s and UploadReport is Programs areCreated Executed

Errors are

Corrected

MainActivitiesin Data Migration

1. Define Data Objects and Attributesto be Migrated2. Configure appMIGRATE Extraction TemplatesAccordingly3. Extract Selected Objects and Attributesfrom

SourceInstanceand move to a StagingTable4. Transform and Cleanse Data as Required5. Pre-Validate Data Prior to Load to eBS TargetInstanceCorrect Any Pre-validation Errors6. Load Data to Target Instance

DefineData Objectsand Attributesto beMigrated

-Ourmethodology beginswith facilitated work shopstodefinedataobjects and attributes to be migrated, anddatavalidation,transformationand cleansing rulesandrequirements-Theworkshopshelpto build stakeholderownership-Thisinformation is then documented and source object /attributeto target object / attribute mapsand data validation/ cleansing rulesarecreated and approved-Thesedocuments arethe basisforthe configuring oftheappMIGRATE templates

•

Configure appMIGRATE Extraction Templates

appMIGRATE features over 200 extraction

•

•

adapters and load templates that include everyobject and attribute in the standard eBS databaseTemplate configuration involves mapping thesource object attributes to the target objectattributes this includes any user definedattributes and additional fields (Definitive FlexFelds)CHAINSYS will be pleased tocomplete theconfigurationtask or totrain client personnel in

configuring the templates

1

2

andmove toa Staging Table• The extract andload templates are selected andthe

sourceinstance and file(s) tobe extracted areidentified. Upon execution of“Run-Loader” programdatais copied from thesource objectto theDataMartand is available for review and edit. Acompleteelectronic record of source location ofeachrecord is maintained

ExtractSelected Objects andAttributes from Source Instance3

•

Cleanse and Transform Data as Required

appMIGRATE providesa robustlogical datatransformationtool. Itis capable oflogically

modifyingexisting data orcreating data logically inpre-definedfields. For exampleif a newsegment isaddedto the general ledger chart set the entry canbecreated using this feature. appMIGRATE includes anumberof tools tosupportdatacleansing, for

example data may be compared to locateduplicate records with alias key identifiersbased on sound, key words (LevinsteinDistance Method) and custom configuredlogical relationships. All objectsand attributesinthedata mart are available foredit. An audittrail iscreatedfor all changes

4

Pre-Validate DataPrior to Load to Target InstanceappMIGRATE will performa pre-validation of the data in thedatamart using the samecriteria asthe target instance. -Inaddition special validation rules maybe added to theappMIGRATE pre-validation if needed. -An on-line errorreportispresented and the usermay makecorrections asrequired-Atool to assistin the correction oferrors(bysuggestion) isalsoprovided. -All data correctiondata is collected andmaintainedand isavailable in an electronic document form

Load Data into Target Instance

appMIGRATE uses standardeBS Application ProgrammingInterfaces(API’s) toensure thatdatais validated inaccordance with the eBS rules. Wherestandard API’sare notprovidedby eBS appusescustom API”s andOpenInterfaceRules to providefordata validation-Load executionsmay be scheduled automatically using FileTransferProtocol(FTP) proceduresstandard to eBS-Load execution isrun asa “background task” permittingotherapplicationsto be usedsimultaneously

6

Oracle E-Business SuiteR12.1.3 Migration

Approach

Data transform and cleansing

Data Transformation

• OPM inventory to Discrete inventorytransformations

• New Structure and accounts for GL accounts• New Item Codification• New Sub-inventory and Locator implementation• Projects / task included in the Inventory, PO and

Sales transactions.• Transformation occurs on the data mart.• Rules Engine for Transformation• Cross References

Data Cleansing

• Manual and Automated cleansing• Threshold value for automated cleansing• Extract the data and load into the Data mart for

cleansing• Manual cleansing by exporting the data mart data

into excel spreadsheet• Automated cleansing using Levenstein’s distance

method, soundex and custom rules.• Audit trail for the cleansed data• Validate the data after cleansing before it is loaded

into Oracle EBS R12.1.3

Data Consolidation

and

cross

references

Collect

Master Datafrom allSources

Standardize/ HarmonizeMaster Data

Review

andCleanseMaster Data

Data CrossReferencing /

HarmonizationFile

ConsolidateData in aCommon

Repository

Collect /Consolidate

TransactionalData from allSources

Standardize /Harmonize

TransactionalData

appLOADData

Repository

eBS Instance 1 eBS Instance 2

J D EdwardsEnterpriseOneInstance

Legacy ERPInstance

e●Chain™Instance

Data Consolidation

and

cross

references

An alternate approachto DataStandardizationis DataHarmonization. In this techniqueall unique records are still linkedvia a commonand standard keyidentifier. A cross reference

table, similar to that used for partnumbercross referencingisdevelopedto cross referenceoriginalkey identifiersfrom thediverse data repositoriesto thestandardkey identifier. Thistechniqueprovides a means ofobtainingthe standard keyidentifierwhen the original keyidentifieris presented.

Key TakeAway – Recommended Solution

Key

Take Away

Thank You.