Product Quality Giveaway Reduction Program Supported …cdn.osisoft.com/corp/en/media/presentations/2014/UsersConference... · Product Quality Giveaway Reduction Program Supported

© Copyr i gh t 2014 O SIs o f t , LLC .

Presented by

Product Quality Giveaway

Reduction Program

Supported by PI System

Zsolt Nagy

Senior Engineer - MOL


Hungary

BudapestCapital

93 028 km2Area

10 millionPopulation

12Nobel prizes

476 (All-time 8th)

Olympic Games medals

2

© Copyr i gh t 2014 O SIs o f t , LLC . 3


PI System Portfolio in MOL

PI ClientsPublish data via

PI ProcessBook

PI DataLink

PI Coresight

PI WebParts

PI SDKDeveloped

applications to support refinery

functions

PI AssetFrameworkCollect data from the

field and create unified asset

hierarchy

PI Notifications

Alerting platform based upon the PI

AF architecture

PI ACEWrite complex

equations, which are reusable for similar

data sets

4


Agenda

About quality giveaway

Specification register

Statistical quality control

Tank quality integration

5


Agenda





6


What is Quality Giveaway?

7


Reducing Quality Giveaway

8


Agenda





9


The Specification Register

Located in NICE launch pad The main screen of the application

10

PI SDK based development



Min. and max. specifications can be

registered

Trending possibility

11


Records minimum and maximum specifications to the PI System to make these data more accessible for …


• PI DataLink

• PI ProcessBook

• PI Coresight

… PI Clients for easier expert examinations

• Advanced process controls (APC)

• KPI system (SEMAFOR)

• Opralog (E-logbook)

… other systems

12


Lab results and specifications are located in the same PI System database

All matched data is gathered in PI Asset Framework for easier examination

Specifications in PI System

Lab result

Minimum specification

Maximum specification

APC calculation (if exists)

On-line analyzer value (if exists)

13


Agenda





14


Statistical Quality Control

Control chart

Statistical quality control (SQC) is a statistical-based method is applied in order to monitor and control

a process to ensure that it operates at its full potential – making conforming product with a minimum of

required energy.

15


The Conception

Creating a range from the examined lab results

This range will represent the process in the examination

Examination covers only key qualities of products

Minimum one sample per each day

Statistical analysis requires 35 independent data

16


Basics of the Examination

Characterizing 5 specifics of the process

Fits in the range

Keeps hard limit

Keeps soft limit

Efficiency

Being controlled

17


Does the Process Fit in the Range?

The statistically created range can be compared with the range defined by the specifications.

Does not fit in the range Fits in the range

18


Does the Process Fit in the Range?

The statistically created range can be compared with the range defined by the specifications.

Does not fit in the range Fits in the range

19


Does the Process Keep Hard Limit?

Lab values are stepping over the hard limit.

Does not keep hard limit Keeps hard limit

20


Does the Process Keep Soft Limit?

Lab values are stepping over the soft limit.

Does not keep soft limit Keeps soft limit

21


Is the Process Efficient?

The location of the statistically created range compared to the range defined by the specifications.

Inefficient Efficient

22


Is the Process Being Controlled?

The standard deviation of the lab results can be compared with the moving standard deviation of the

lab results (stdev of all lab results vs. average of the stdev of all consecutive lab result pairs).

Not being controlled Being controlled

23


Process in Operation Example 1

Fits in the range

Keeps hard limit

Does not keep soft limit

Inefficient

Being controlled

24



Fits in the range

Keeps hard limit

Keeps soft limit

Inefficient

Not being controlled

25



Fits in the range

Keeps hard limit

Keeps soft limit

Efficient

Being controlled

26


Agenda





27


ProblemInappropriate place of sampling (POS) for qualification

Temporary violation of limits in in-line analyzer can be allowed

Product quality in tank is important, it is more robust to loads

28


SolutionThe solution is to calculate the quality in the tank

Setpoints are determined in order to fulfill the specifications in the tank, not at the end of the pipe

29


Technical Solution of the Problem

In-line measurement is available in the PI System

TQI is just a question of calculation in many plants

Wide spectrum of opportunities

Possible ways of realization

Spreadsheet with PI DataLink for pilot tests and for research activities

Integration of TQI in APCs or DCSs

(PI ACE,…)

30


Benefits of TQI

Unnecessary give-away can be eliminated

• Laboratory analysis is slower and less often available

Information about product quality is quickly availabe any time

• Disturbances in measurements are eliminated by weighted averaging

• Place of sampling is indifferent

Ensures smaller variance than the ISO standards for laboratory measurements*

Provides data for further optimization

31


Identified Improvement Potentials, Examples

DCDU2 unit white product yield can be improved via

VGO T95

Potential loss: 8,5 M USD/year

(1,2% white product yield)

DCDU 3 unit white product yield can be improved via

VGO T95

Potential loss: 6,9 M USD/year

(0,6% white product yield)

32


Future Plans

33


Roll-out to other sites

SQC

Project initiation in

2015

TQI

Project initiation in

2015


Zsolt Nagy

[email protected]

H-2443

Százhalombatta

P.O.Box. 1.

34

mailto:[email protected]


Brought to you by

Product Quality Giveaway Reduction Program Supported …cdn.osisoft.com/corp/en/media/presentations/2014/UsersConference... · Product Quality Giveaway Reduction Program Supported

Documents