|
|
|
|
|
  
|
REDUCT APPROACH TO EFFECTIVE PROCESS MANAGEMENT
|
||
 Services
Papers:
|
Background Modern
process industries require timely information and data to meet the challenges of
today’s fast changing business environment. Management
can meet these challenges and competitive forces by having the ability to respond with the
right strategies. This requires reliable
process data, as well as the ability to make sound decisions. Organizational management needs, therefore,
effective tools to process a variety of information, and to identify and implement timely
actions. Today’s
plant information systems are very effective in collecting, summarizing, interpreting and
displaying process data. The problem is,
however, that there is often too much information: the amount of data is simply
overwhelming. What is needed, therefore, is
not more data, but more “process intelligence”. Process Intelligence
Process
Intelligence (PI) is a code-name for a range of technologies which unlock knowledge hidden
in process data while telling how to use it more effectively. The overall objectives of PI are to improve
productivity, product and service quality, and business profitability by making
information and data more comprehensible and using process knowledge more effectively. The
specific objectives are to leverage investments in instrumentation, PLC’s, SCADA and
DCS systems, and plant information networks through better process intelligence. In short,
Process Intelligence helps to use process data for
competitive advantage. PI assists
process engineers, plant superintendents and managers in making everyday decisions about
processes. It uses intelligent software technologies for better, more optimal reasoning
about process data. PI technologies apply
sophisticated algorithmic methods based on advanced statistics, operational research,
Artificial Intelligence, evolutionary algorithms, etc., for tasks such as data analysis,
system control and optimization, knowledge acquisition, or adaptive scheduling. Process data can be gathered, interpreted and made
available in real-time throughout the plant using Internet Web technologies, OLE for
Process Control (OPC), On-Line Analytical Processing (OLAP), MS COM/DCOM, etc. Systems integration technologies
focus on integrating various equipment control functions, and information services focus
on implementing business transactions. Process Intelligence, in contrast, focuses on
process data, process information, knowledge, and all other business and operation aspects
involved in scheduling , optimization, control, maintenance and management of the process. Process Intelligence provides tools to help
process engineers and managers to make better process decisions.
Process Intelligence is a "bridge" between the machines' management and the business management:
Process Intelligence can:
The specific capabilities of Process Intelligence technologies are: OLE for Process Control (OPC) allows access to process
information in DCS, SCADA and PLC’s in a standardized way, thus enabling open and
more efficient communication in a heterogeneous computing environment. OPC offers an
interface with speeds up to 10,000 points per second, high robustness and reliability, and
advanced security on PC computers. OPC client applications can be built at half the cost
of proprietary system architectures Data mining enables understanding of complex
processes and issues where the engineering and scientific principles are not well
understood. It discovers how and why process
experts and operators make decisions under diverse circumstances. The user can also learn about the complex
strategies used by the best shift teams by mining their historical production records, and
discovering how they reacted to various production events. Advanced algorithm methods like Genetic Algorithms build real-time, adaptive scheduling systems that incorporate process economics into production schedules. They allow elimination of inefficiencies in utilizing production capacity by integrating inventory, current orders and production plans. These methods facilitate easy rescheduling, review of alternative schedules and quick assigning of production orders. Virtual On-line Analyzers using technologies such as neural
networks to model and predict process measurements that are difficult or costly to
perform. They act as “virtual”
instrumentation or sensors and can be easily updated as equipment ages, retrofits are
done, or process operating characteristics change. They
also help to explore the relative impact of process variables through “what-if”
analysis and to analyze process economics. Supervisory DecisionExperts perform multiple tasks in
supervisory control, optimization and automation. They
help evaluate and implement optimization strategies based on process knowledge discovered
from process data. The operator can address
varying objectives and constraints simultaneously while compensating for changing feed
conditions and process dynamics. Since there
is no need to develop process models, DecisionExperts are less costly and more profitable
than other methods. Intranet Web technologies make process
information available plant-wide in a format that addresses employees’ needs. Whether it is the process accountant, process
expert, or production manager, each receives up-to-date process information that is
prepared for easy use. Business Intelligence using technologies like
OLAP allows management to gain insight into their business by “drilling-down”
business data for trend spotting or using “what-if” analysis for scenario
forecasting. BI technologies provide
management with up-to-date information on vital business aspects of their operations.
Risks and Challenges
Implementing
Process Intelligence requires some changes to how an organization works with data.
These changes will create a variety of reactions including disbelief, skepticism,
hesitation, or fear that a PI project will fail and reflect negatively on its champions.
There is no simple answer to every possible fear of change, but there are effective
ways to manage the risks and challenges associated with PI projects.
The foundation of PI risk management
is a two- phase approach to its implementation: First a feasibility study is conducted
where benefits and risks are identified and evaluated with actual process data. Then in a second stage, selected opportunities are
implemented after all questions are resolved. The
risk that a PI project may miss its targets and objectives is reduced by the formation of
a project team consisting of company process engineers and management, as well as
knowledgeable consultants. This team will
manage all technical risks and staff training, and keep company management abreast of all
significant milestones. Finally, the possibility that a PI
project will divert a large amount of staff time away from everyday urgent tasks is
remedied by having consultants perform most of the detail project work, thus providing
additional, temporary manpower and capabilities to the company management.
Summary Both competitive business pressures
and the uncontrolled growth of process data have contributed to the need for technologies
that can assist process engineers and managers in dealing with the information overload
and in using data more effectively for competitive advantage. Process Intelligence offers such technologies
today. Process Intelligence systems provide vital business and process intelligence to
industry, and PI technologies are within the reach of every plant.
|
||
| REDUCT & Lobbe
Technologies Inc. P.O. Box 800, 186 - 8120 No.2 Road., Richmond, BC, Canada V7C 5J8 ph: (604) 275-3711 fax: (604) 275-3715 email: dispatch@reduct.com |
Process
Management