|
The Ultramax Corporation (UMC) was
inspired by the principle that a worthwhile strategy in production is not
only to generate value, but to continually learn how to do it better (an
original idea of Box & Draper). In fact, with the right tools, it is a
waste to do anything less.
Further, continuous improvements, especially when aided by serious
optimization, in addition to being profitable it is exciting and fun,
contributing to the quality of life of all participants. The incremented ability of the plant
personnel makes the business capable of coping with future changes,
challenges and opportunities much better. Most existing production processes have
the potential to significantly increase their contribution to company margins
simply by adjusting better control settings at which to operate. Valuable performance improvements may
come from combinations of better quality, higher production rates, reduced
material usage, less waste, lower energy consumption, reduced emissions and
other benefits that directly impact profits. While a process may perform
adequately, there is usually opportunity for a great deal of improvement
resulting in cost savings and higher profits. Obtaining this performance
improvement has been a very difficult task with the tools available in the
past. The early development of the UMC
empirical technology for process optimization was driven by the need to find
a low-risk and efficient method to improve existing processes, during
production, without intruding on normal operations. Tools then in use such as
Evolutionary Operations (EVOP), SIMPLEX, and Design of Experiments (DOE) all
depended on experimental methods that were not well suited for the production
environment. So does the newer
technology Neural Networks. They
either upset the process or took too much time and attention to be
feasible. In addition, because
managers had production goals to meet at targeted costs, they were reluctant
to permit experimentation that interfered with their schedules. To improve process performance,
manufacturing managers needed a product that could advise control adjustments
to continuously improve the process and respond to changing conditions in
order to extract best performance at all times. The challenge was to create a
product that would be able to learn from the process as it ran, discover
better settings and be easy to use.
The idea and applications were started at Procter & Gamble, UMC has been the leader in supplying
its proven empirical optimization solutions to industry since 1982. In 1988 it signed a license agreement
to optimize the production of semiconductors with IBM. Since that time other industrial
giants such as Procter & Gamble, GE, Ford Motor Co., Textron Automotive,
General Motors, Merck, Kimberly Clark, BASF and many others have used
ULTRAMAX to assist them in solving manufacturing related problems and improve
production effectiveness. In 1993, ULTRAMAX launched its first
industry focused marketing effort. This effort was driven by the Clean Air
Act Amendments of 1990 and the resultant challenge to reduce air emissions it
placed on public utilities in the Thereafter ULTRAMAX consistently
demonstrated its effectiveness in optimizing boiler combustion to reduce NOx
and meet the clean air standards set by the EPA as well as gain improvements
in heat rate and LOI. Because
ULTRAMAX has the unique capability to be used without direct interface to a
control system (stand-alone mode), many early projects were performed in this
way. Often, the project objectives were to deal with special problems to meet
NOx guarantees, avoid de-rating or demonstrate effectiveness on a different
type of boiler. In every case,
ULTRAMAX showed that plant operators could expect to obtain NOx reductions of
10% to 40% and lower heat rate by 0.5 to 3%, without negatively affecting
other key operating requirements. These applications proved that valuable
improvements were possible from software solutions before the capital
investment of hardware additions. As acceptance of software optimization
solutions increased, generation plants began to see the value in interfacing
ULTRAMAX with their control systems to enable continuous performance
optimization for all load levels and with changing conditions due to fuel,
environmental conditions and gradual degradation. The new competitive nature of the
power industry brought on by deregulation has put increasing demands on plants
to obtain least-cost compliant operation. Utilities have come to think of
ULTRAMAX as an integral part of their overall NOx compliance and performance
improvement strategy in addition to low-NOx burners, overfire air,
re-burning, SCR, SNCR and other boiler modifications. The operation of these hardware
additions as well as plant components such as precipitators, mills and
scrubbers can also be improved using ULTRAMAX. ULTRAMAX is part of an integrated
solution that combines these various processes to obtain optimized operation
throughout the plant. In 2002 a
NN competitor acquired the exclusive rights to market, sell and service
ULTRAMAX in the power industry in the UMC delivers this integrated solution
in closed-loop supervisory or operator advisory modes. The Closed-Loop format
changes control settings automatically without operator intervention –
and returns to Advisory when an upset or undesirable operations are detected
for immediate plant personnel attention. The Advisory system format allows
the operator to review recommendations and make the final choice before
control settings are changed. For cases where external demands or
conditions keep on changing, such as for a load-following boiler, UMC created
a more powerful solution to optimize on top of the transients so created:
"transient optimization". UMC continues to also provide solutions
to companies in the chemical process, automotive, aerospace, semiconductor,
consumer products and plastic forming industries. Key targeted industries are processed
foods and metals. Its software products improve the value of continuous,
batch and discrete processes while applied in closed-loop, advisory and
stand-alone modes. |
|