Framing Quality with PI Data - etouches · 2007 - User Conference - Monterey NWA Statistical...

Presented by

© Copyr igh t 2013 OSIso f t , LLC Pac i f i c A lumin ium, Ke i th S inc la i r and Assoc ia tes .

Framing Quality with PI Data

Chris Noonan


Agenda

Pacific Aluminium at a Glance (10 min)

Our Journey with OSIsoft (10 min)

Quality Framework (13 min)

Application of the Quality Framework (10 min)

Wrap Up and Questions (2 min)

2

9:45

© Copyr igh t 2013 OSIso f t , LLC, Pac i f i c A lumin ium, Ke i th S inc la i r and Assoc ia tes . 4

Full Version http://www.youtube.com/watch?v=uU_ldmWyoEs


Pacific Aluminium at a Glance

Our Journey with OSIsoft

Quality Framework

Application of the Quality Framework

Wrap up and Questions

5

9:55

© Copyr igh t 2013 OSIso f t , LLC, Pac i f i c A lumin ium, Ke i th S inc la i r and Assoc ia tes .

2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

6

2012: PIMS Quality Framework Delivered to NZAS Metal Products

2000 • IS&T Strategy

recognised real time data was being lost and was unable to be analysed

• OEE Project started

2002: OSIsoftSelected

2002: Pilot at Bell Bay Cast House established basic design and principles (PI ProcessBook, PI Batch and PI Module Database)

2003->2004: BSL and NZAS Cast HousedProve design transportable Established a permanent team – RTA and contractors

2003 ->2007 Configurable Data Entry developed.

2005->2007: Carbon plants are reaching the limitations of the technology without the relational capabilities of AF and Event Frames

2007->2008: Overhead cranes - Focus on operational and maintenance data

2000 Rio Tinto buys Comalco

October 2007 Rio Tinto buys Alcan

“Integrated Process Management System” –Sinclair Associates

March 2013 IT Systems Separate

October 2011 Pacific Aluminium Formed

2006 XML / XSL Asset Templates “Standard Objects”

2006 -> Statistical Process Control UI Developed

2005 Migration to PI ACE .NET. Batch Factories and Batch Monitors

2013 Rationalisation and configuration of PIMS calculations

2008 -2011 Global Focus 2011 Regional/ Local Focus

2007 - User Conference -Monterey

NWA Statistical Process Control

2009 PI AF 2.02011 - User Conference –San Francisco

2006 - User Conference –San Francisco

RTA holds Metals and Mining Sig Leadership

2012 Event Frames

2013 Publish Subscribe Events Architecture

Our Journey with OSIsoftStop

Press!


Tags out of the BoxShow what drives this measure

Show with time context

Reuse on similar plant

Find data for assets easily

Quickly get at Data (if you know the tag name)

7

Process Book or Datalink• Link directly to the tag –no abstraction

Historian• Tag (e.g. FURNA_TEMP13_PV”) with time series data from the process


With an Asset Model

8

Process Mimics•Link indirectly to the tag via the alias

Asset Framework•(Furnace A – Roof Temperature)

Historian•Tag (e.g. FURNA_TEMP13_PV”) with time series data from the process

Show what drives this measure and how it is managed




Quickly get at Data


BakeOEE

Alarms

Mixer Alarms

Module Database Asset Templates in 2006

XML XSL Module DataBase

9


Plant

Green Carbon Former OEE

Bake OEE

Plant Green Carbon Mixer Alarms

OEE Filter

Asset Filtering

10


Asset Templates in 2013?

11

• The “PI Asset Framework (AF)”


Add a Time Model

12

Historian (PI)

Asset Framework

Process Mimics

Event Frames

OEE , BatchShow what drives this measure and how it is managed




Quickly get at Data


Framing process data in “Event Frames”

13


Add a Quality Framework Model

14

Historian (PI)

Asset Framework

ProcessBook

Event Frames

OEE , Batch

Quality Framework

Quality Framework

Presentation Layer

Show what drives this measure and how it is managed




Quickly get at Data


Dimensions

15

Quality

People

Time

Assets




Quality Framework



16

10:05


Why Frame Quality?

17

• Desire to see “personalized” information on their screen (not having to drill down through an asset structure”

• Quality Systems are still embedded after 13 years, but need support to lock in business knowledge.

• Fewer people need more information pre-packaged to make the right decisions faster

• Appetite for practical problem solving


Your First Day on the job and…… A customer complains about the product you are sending them. Where do you look in the asset structure for answers? Which tag?

The California Governor reviews an OSIsoft ProcessBook Display when the California ISO Reported a New High for Peak Power Demand

18


Quality Maturity Model

19

• Processes at this level that the focus is on continually improving process performance through both incremental and innovative technological changes/improvements.Level 5 - Optimizing

• Processes at this level that, using process metrics, management can effectively control the AS-IS process. In particular, management can identify ways to adjust and adapt the process to particular projects without measurable losses of quality or deviations from specifications. Process Capability is established from this level.

Level 4 - Managed

• Processes at this level that there are sets of defined and documented standard processes established and subject to some degree of improvement over time. These standard processes are in place (i.e., they are the AS-IS processes) and used to establish consistency of process performance across the organization.

Level 3 - Defined

• Some processes are repeatable, possibly with consistent results. Process discipline is unlikely to be rigorous, but where it exists it may help to ensure that existing processes are maintained during times of stress.Level 2 - Repeatable

• Processes are (typically) undocumented and in a state of dynamic change, tending to be driven in an ad hoc, uncontrolled and reactive manner by users or events. This provides a chaotic or unstable environment for the processes.

Level 1 - Initial (Chaotic)


How do we move up the Maturity Model?

20

Understand Cause and Effect (Relationships)

Let the process tell us when not in control or capable

(SQC)

Understand the what the

customer is willing to pay for

(What’s important)

Lock in agreed action plans and accountabilities

(Plan)


What do Customers Want?

21

• My Aeroplane motor doesn’t ceaseOil and Gas

• Electrical losses in my cable are lowMining and Metals

• My toaster doesn’t diePower Generation and Transmission

• My printer doesn’t jamPulp and Paper

• I don’t get sickFood and Beverage

• The side effects are minimalPharmaceuticals

Understand the what the

customer is willing to pay for

(What’s important)


Relationships

22

Understand Cause and Effect (Relationships)

Customer Requirements

Product Properties

Properties at the end of each process

Critical Process Variables

Maintenance and Procedures

© Copyr igh t 2013 OSIso f t , LLC, Pac i f i c A lumin ium, Ke i th S inc la i r and Assoc ia tes . 23

Leveraging the PI SQC TagLet the process tell us when not

in control or capable (SQC)


Leveraging the PI SQC Tag

24

• Leverage Product• Store Control Limits• Store Capability

Limits• Store Comments• Change Aggregation

Methods

Let the process tell us when not

in control or capable (SQC)


Oil Analysis Response Plan Example

25

High copper is found in the oil sample

Two maintainers change out the gearbox, strip it down and prepare to repair it.

Response



Oil Analysis Response Plan Example

26

If a response plan was in place for high copper, would the maintainers have changed out the gearbox and not the pump which contained brass?



Locking in Improvement

27

Measurement Plan• How will the measurement be taken• Who and How Frequent?

Control Plan• How will the control charts be displayed• Who and How Frequent?

Response Plan• How will we respond in a standard way• Who?



The Quality Framework Model

28

Role

Measurement, Control, Response Plan

Person

Organizational Hierarchy

Frequency

Relationship to upstream and

downstream metrics

Measurement

PI Process Data

Customer – Supplier Relationship

Relationship type

Relationship Strength (H,M,L)

Standards

Process changes

Actions

Metric


Show me all my measures that are not in control

29

Role


Person


Frequency


downstream metrics

Measurement

PI Process Data


Relationship type


Standards

Process changes

Actions

Metric


Show upstream measures

30

Role


Person


Frequency


downstream metrics

Measurement

PI Process Data


Relationship type


Standards

Process changes

Actions

Metric


Compare FurnaceTemperatures

31

Role


Person


Frequency


downstream metrics

Measurement

PI Process Data


Relationship type


Standards

Process changes

Actions

Metric


Compare “Standards”

32


What outstanding actions are there for my team?

33

Role


Person


Frequency


downstream metrics

Measurement

PI Process Data


Relationship type


Standards

Process changes

Actions

Metric


Actions

34


Impact of upstream process changes (Ball Mill)?

35

Role


Person


Frequency


downstream metrics

Measurement

PI Process Data


Relationship type


Standards

Process changes

Actions

Metric


Finding Data

36

“…it starts to make sense of the wealth of information that is buried within PIMS.”

“It’s not that the data is not valuable it’s more that the critical few are not presented/displayed in the most helpful way. The Quality Framework would appear to provide a method for solving this issue – presenting data in a helpful way.”

“To the vast majority of people who are not “über-users”, finding your way through mountains of PIMS data to find what you need can be challenging.”

© Copyr igh t 2013 OSIso f t , LLC Pac i f i c A lumin ium, Ke i th S inc la i r and Assoc ia tes . 37

Standards

Documentation

Accountabilities (who and when)

RelationshipsActions

Process Changes




Quality Framework


Wrap up

38

10:20


The Problem – High Scrap

39

“All those involved on the billet process scrap reduction project have seen the potential power of the framework and devoted significant time to its construction – primarily as it makes data analysis far more efficient and effectively provides the Control step of the DMAIC process.”

“Controlling Scrap is worth $3,200,000 per annum”

“One of their first challenges has been to stabilise the process and reduce variation. This has required the charting of the CPVs and development of response plans. The Quality Framework has been the ideal solution to achieve both activities.”


What was required

40

Cultural change and belief that reducing variation will reduce costs.

A coach to help change the culture and coach variation reduction

techniquesSupport from a sponsor with a process not “in

control”The Quality Framework

System

Timely, Accurate Data


Examples

41


Supporting ISO9001 Certification

42

“The work that Chris and Melissa did last week in mapping the process and the codification of the output within the Quality Framework will tick this box very nicely. Once implemented, the Quality Framework should virtually eliminate the need to repeatedly keep documentation current etc. This is a reasonable resource saving (Melissa will thank you for it!!).”

“The key recommendation that the auditor gave was to ditch our traditional quality manual (linked to an older version for the ISO9001 standard) and adopt a layered descriptive “process flow” approach.”

“In effect this changes ISO accreditation from being an “add-on”, with the attendant administrative burdens to keep documentation current, to a valuable training and communication tool that reflects the process and its interdependencies.”


User Feedback

43

“Placing cycle time data within the Quality Framework will give us the ability to make cycle time reduction decisions efficiently and robustly.”

Instead of using SPC we have substituted it with “have we got a problem or not with this process?” As you can appreciate, we sometimes do not respond when we should, and other times we respond to process noise. We anticipate that the use of SPC on our process metrics will greatly improve the deployment of our scarce resources – only responding to process when it actually tells us something (actioning special cause exceptions and pre-empting out-of-control).

© Copyr igh t 2013 OSIso f t , LLC.

• Scrap rates continue to drop, the cause and effect is now being locked in to the corporate knowledge.

Solution Results and Benefits

Pacific Aluminium: Framing Quality with PI Data

Business Challenge• High scrap rates in the

Billet plant was costing the business a lot of money

• Pulling PI Data together with a meta data framework enables people to find the right data quickly.

“To the vast majority of people who are not “über-users”, finding your way through mountains of PIMS data to find what you need can be challenging.

It’s not that the data is not valuable it’s more that the critical few are not presented/displayed in the most helpful way. The Quality Framework provides a method for solving this issue – presenting data in a helpful way.”

44

10:30




Quality Framework



45

Questions?


Chris [email protected] Value Realisation & Bus. Intelligence, Production Systems Pacific Aluminium

[email protected]

46

Keith Sinclair



Quality Framework



Brought to you by


Sinclair Associates

Framing Quality with PI Data - etouches · 2007 - User Conference - Monterey NWA Statistical...

Documents

Transcript of Framing Quality with PI Data - etouches · 2007 - User Conference - Monterey NWA Statistical...