PETRONAS E&P Technical Data Quality Metrics Initiative – Going Green with Data

Hassan B SabirinRegulatory Compliance & Technical Assurance, Technical Data

Hassan B Sabirin, 8 October 2013, Page 1

Regulatory Compliance & Technical Assurance, Technical Data

Petroleum Management Unit (PMU)

PETRONAS E&P

08 October 2013Impiana Hotel, Kuala Lumpur

Digital Energy Conference, Kuala Lumpur:Improving E&P Subsurface Data Management

© 2013 PETROLIAM NASIONAL BERHAD (PETRONAS)

All rights reserved. No part of this document may be reproduced, stored in a retrieval

system or transmitted in any form or by any means (electronic, mechanical,

photocopying, recording or otherwise) without the permission of the copyright

owner.

Contents

• Organization Background• PETRONAS EP Technical Data

• The role of data management in value creation

• What problems are we trying to solve?

• Data quality metrics defined

Hassan B Sabirin, 8 October 2013, Page 2

• Example – Some basic metrics

• Data quality metrics – high level going green roadmap

• The data quality metrics and standards connection

• Concluding remarks

Technical Data was formed to elevate functional excellence and delivery discipline of data management

PETROLEUMMANAGEMENT

PETROLEUMRESOURCE

EXPLORATION

PETROLEUMRESOURCE

DEVELOPMENT

PETROLEUMOPERATIONSMANAGEMENT

STRATEGICPLANNING

FINANCE & ACCOUNTS

PSC SCMGOVERNANCE

HSETECHNICAL

DATA

Hassan B Sabirin, 8 October 2013, Page 3

Knowledge Management

& Institutional Capability

Institutional

Capability

Knowledge

Management

Business Process

Advisory &

Solution

Planning &

Performance

Program

Excellence

Data Facility

Management

Regulatory

Compliance

Technical

Assurance

Geospatial

Geophysical

& Geological

Application

& Workflow

Project Data

Application

& Workflow

Production

Technology

Petrophysics

Reservoir

Engineering

Application

& Workflow

Facility

Engineering

Well Engineering

Process Data

Application

& Workflow

Planning&

Delivery

RegulatoryCompliance

& Technical Assurance

GeoscienceData

Project Data

PetroleumEngineering

Data

Production&

OperationData

Data Quality Metrics

TD

PETRONAS MANAGEMENT

PSCs / RSCs

3rd Parties

Government Depts

Industry Standards Peer

EP

Business

Strategy External Data

Benchmarking

& Request Data

Consultancy

Services

Technical Data Department Relationship Model

PETRONAS

Hassan B Sabirin, 8 October 2013, Page 4

TD

EP IT

Industry Standards Peer

Groups

Technical Apps &

Data Vendors

Consulting Services

ICT &

Infra

Host Authority

Data

Requirement

DM

Services

PETRONAS

EP Business

Units

> 1,000,000 media

> 700,000 technical

>766 active projects

~ 159 ventures in 25

countries

~ Average 100 wells

drilled per year

~ 1,000 2D/3D seismic

surveys

~ 12,000 wells

- All EP data types

>250 specialized

technical applications

(Multiple disciplines &

Measured Data> RM35 Billion investment on

data acquisition

Interpreted (Result) Data

Subsurface understanding which has direct

impact on business decisions

Dimensions of

Background – PETRONAS EP Data Dimensions

Hassan B Sabirin, 8 October 2013, Page 5

> 700,000 technical

docs

> 3,500 TB disk space

>120 data bases

Tools (Technical Software Application)

> RM1.2 Billion investment

(Multiple disciplines &

related application

workflows )

End Users> 400 geoscientists and engineers from all

technical domains in E&P

End Users> 400 geoscientists and engineers from all

technical domains in E&P

Media & Corporate DBsMedia, hardcopies and Geosamples

External StakeholdersPotential investors, PSCs, government

agencies, universities

External StakeholdersPotential investors, PSCs, government

agencies, universities

Dimensions of

PETRONAS EP

Data

Management

Workflow and Standardization

Project

Database

ApplicationSource / feeder

databases

Master (Reference) Data

Data Quality Metrics

Corporate

Database

Publishing

Previous studies : Geologists spend up to 60% of their time looking for & quality checking data for their projects

Re-useValueLoop

Corporate Data

2 2

3

3

unique master sourceof quality checked dataunique master sourceof quality checked data

Role of Data Management in Value Creation

Hassan B Sabirin, 8 October 2013, Page 6

Data

collation

Project set up

Database Database

Publishing

Re-use

ValueLoop

Cost management

Process improvement

Optimization

Reduce the 60% to __%

Standards & workflowsBring to acceptable qualityCapture work done for re-use

‘One version of truth’ datasetsRe-use many times

1

1

1

2

2

3

3

Approved & final Approved & final

Feeding new projectsFeeding new projects

The project space is where the big decisions are made

Reduction of time for data collation and project setup

Step 1

Step 2

Step 3

What Problems Are We Trying to Solve?

1- Mitigate business risks created by data

quality issues

• Financial

• Confidence/Satisfaction

• Productivity

Hassan B Sabirin, 8 October 2013, Page 7

• Productivity

• Compliance/Legal

• Safety

What Problems Are We Trying to Solve?

2 - Value leakage in business

• Lost time due to re-checking of data

• Not re-using data because of trust issues

• Lack of awareness & enforcement on standards –

inconsistent business practice

Hassan B Sabirin, 8 October 2013, Page 8

inconsistent business practice

• You can’t manage what you can’t measure – quantification of

leakage

What Problems Are We Trying to Solve?

3 - Managing change introduced by new technology

New data content may introduce a requirement to improve

• Data standards/guideline

• Data integration process

Hassan B Sabirin, 8 October 2013, Page 9

• Data integration process

• Data analytics method

Data Quality Metrics Defined

• Data Quality Metrics: A measure of data quality performance

• Data Quality Dimensions: Perspectives from which data quality is being looked at

Hassan B Sabirin, 8 October 2013, Page 10

• Business Rules: Rules that must be adhered to in order to keep the business running in good/optimum condition

Project A

Data Quality Metrics Defined

Completeness

Validity

Deviation Survey

Data Type Quality Dimension Metrics Target

Data Quality Metric

Hassan B Sabirin, 8 October 2013, Page 11

Validity

Uniqueness

Conformity

Consistency

Integrity

Database B

Business Rules:

Well/wellbores with a

completion date 3

months back from

today must have a

definitive/final

deviation survey

Data Quality Metric

Total Error Total Records

)%1 – (

Metrics

Well measured depth must be more or equal than true vertical depth

Offshore well must have a water depth information

A well deviation header must have end/bottom depth of the survey defined

A well deviation header must have start/top depth of the survey defined

A well deviation header must have the survey contractor/company performing the job defined

A well deviation header must have the survey tool name defined

Borehole survey

depth range

Borehole

Examples of Data Quality Metrics

Well Header

Hassan B Sabirin, 8 October 2013, Page 12

A well deviation header must have the survey tool name defined

A well deviation survey point must be identifiable to a deviation survey header

A well deviation survey point must have the calculated dog leg defined

A well deviation survey point must have the calculated vertical section defined

A well deviation survey point must record the measured azimuth

A well deviation survey point must record the measured inclination

Borehole

survey tool

identity

Borehole

survey

measurements

Data Quality Metrics Aggregation

Metric/QuerySurvey must have at

least azimuth,

inclination, MD and

TVD Metric/Query Well Deviation Survey

Metric

Target: Drilling

Database

Borehole Survey

Business RulesAggregation

Hassan B Sabirin, 8 October 2013, Page 13

TVD Metric/Query

Metric/QuerySurvey dogleg must

not be more than 10

degrees Metric/Query

Metric

Queries, Query Sets, Targets & Aggregation

Single Query

Single Query

Single Query

Single Query

Query Set

OPENWORKSPROJECTOPENWORKS

PROJECT

OPENWORKSPROJECT

OPENWORKS

PROJECT

OTHER

DataBasesOTHER

DataBasesOTHER

DataBases

Single Query

Single Query

Single Query

Single Query

Query Set

OPENWORKSPROJECTOPENWORKS

PROJECT

OPENWORKSPROJECT

OPENWORKS

PROJECT

OTHER

DataBasesOTHER

DataBasesOTHER

DataBases

Single Query

Single Query

Single Query

Single Query

Query Set

OPENWORKSPROJECTOPENWORKS

PROJECT

OPENWORKSPROJECT

OPENWORKS

PROJECT

OTHER

DataBasesOTHER

DataBasesOTHER

DataBases

Multiple Query Multiple Query Multiple Query Multiple Query

SetsSetsSetsSets

Multiple

Target SetsSingle

Metric

A G

G R

E G

A T

I O

N MULTIPLE QUERY SETS MULTIPLE TARGET SETS

Hassan B Sabirin, 8 October 2013, Page 14 14

Single Query

eg. select well_name from well_master;

PROJECT

DATABASE

TARGETTARGETTARGETTARGET

Single Query

Single Query

Single Query

Single Query

Query

Set

PROJECT

DATABASEPROJECT

DATABASE

PROJECT

DATABASE

PROJECT

DATABASE

OTHER

DataBasesOTHER

DataBasesOTHER

DataBases

MULTIPLE QUERIES MULTIPLE TARGETS

Single

Metric

Single

Metric

A G

G R

E G

A T

I O

N

Data Quality Metrics Aggregation to Dashboard

A B C D E F etc.Regions

of Interest

The (Enterprise) Dashboard

Bu

sin

ess F

ocu

s

The (Enterprise) Data Quality Trend

Hassan B Sabirin, 8 October 2013, Page 15 15

Databases

of Interest

Data

of Interest

Data Element

of Interest

Assets

of InterestBu

sin

ess F

ocu

sD

ata

Fo

cu

s

Data Quality Metrics Dashboard

Metric

A B

Metric1

2

3

Metric

Metric

Metric

Metric

Metric

Metric

Metric

C

Hassan B Sabirin, 8 October 2013, Page 16

3 Metric Metric Metric

GoodMediumPoor

95-10090-950-90 Total Error Total Records

)%1 – (

Data Quality Metrics Dashboard

Hassan B Sabirin, 8 October 2013, Page 17

Data Quality Metrics Tool

Hassan B Sabirin, 8 October 2013, Page 18

Data Quality Metrics Dashboard – By Drilling Rig

Hassan B Sabirin, 8 October 2013, Page 19

The Data Quality Metrics – Data Standards Connection

Business Rules Data Standards

Data Quality Metrics

measure adherence to

Hassan B Sabirin, 8 October 2013, Page 20

measure adherence to

data standards

Metric

A B

Metric1

2

3

Metric

Metric

Metric

Metric

Metric

Metric

Metric

C

Data Quality Metrics- High Level Roadmap

2013 2014 2015

Infrastructure

ArchitectureArchitecture

Data Quality Focal Person

Da

ta Q

ua

lity

Me

tric

sD

ata

Qu

ali

ty M

etr

ics

Top 4 data types

Corrective Action

Measure Validity,

Conformity

Measure Validity,

Conformity

Measure Completeness Measure Spatial Layers

Top 8 data types

Hassan B Sabirin, 8 October 2013, Page 21

Data Ownership – Roles & Responsibilities

Da

ta Q

ua

lity

Me

tric

sD

ata

Qu

ali

ty M

etr

ics

Oth

er

Are

as

Oth

er

Are

as

Data Standards

Business Rules

Management Support

Project Quality Metrics

Conclusion

•Data Quality Metrics is an important part of the data

management portfolio.

•A data quality metrics program needs to closely

engage business lines• Business must play a role to define the business rules that adds

value to the data.

• Data Ownership – roles and responsibilities must be clearly

Hassan B Sabirin, 8 October 2013, Page 22

• Data Ownership – roles and responsibilities must be clearly

defined

• Proactive participation at the working level, and drive from the top

•Data Quality Metrics results have to tie to business owners

as a measure of performance

Hassan B Sabirin, 8 October 2013, Page 23

Thank You

Paper Abstract

People, process & technology are often touted as the 3 important components an organization must have in order to ensure that the data is managed properly and effectively. While that is true to a large degree, the existence of this tripartite in

the organization does not necessarily support the sustainability of good data management practices over the longer term. Data quality today still remains as a

significant challenge we face in the E&P industry.

For the data we use within upstream E&P, data quality metrics, when implemented

Hassan B Sabirin, 8 October 2013, Page 24

For the data we use within upstream E&P, data quality metrics, when implemented in the right databases, creates a transparency that cannot be easily achieved by any other means, and provides us with a powerful tool to highlight issues before the problem becomes too big. A proper data quality framework also address data issues in holistic way and combines with data ownerships and accountabilities to

ensure the right interventions and corrections get done.

This paper describes PETRONAS E&P’s data quality metrics initiative that will progressively improve the quality of data used by the business and result in a

traffic-light format data quality dashboard that gets more and more green.

Philip Lesslar, 20thMay 2013, Page 2