Operationally challenged, geometrically solved

 
 

Applications                      

The foundation for GPC is a new ability to see and explore an entire process of many hundreds of process variables on one screen and to go on to find better operating limits, Consistent Alarm Limits and Alerts, Best Operating Envelopes without any need for mathematical skills. The underlying objective is to reduce operational variability but in a way that everyone in your plant can understand and will support. It is very rare to find truly novel technological solutions to existing problems so take time to look around and see some of our industry case studies showing major process improvement. Then if you are interested and you want to know more please contact us.
  

Take a look around and discover how GPC
technology products can lead to process improvement for your business.
 
Contact us for more information. 

GPC has been used with outstanding and often surprising results for:

  • Problem solving and process stewardship - addressing the needs of front-line process operations engineers for a quick and easy-to-use method to answer the everyday questions in a plant such as 'why isn't the process working as well as it did last week?' as well as checking that improvements already made remain made.

  • Alarm Management.  It seems obvious when you think about it that all the problems of alarm systems begin and end with the values at which the alarm limits are set. But there has never been a method to calculate them. Now there is. GPC is the only method with a scientific basis for calculating Alarm Limits and Alerts. We do it for many variables at once so it is much faster and the alarm limits are better because they are consistent with each other. Users have found it cuts the time for Alarm Rationalisation projects by at least 20% as well as giving better alarms and a safer process. It makes the subsequent Management of Change easier too.

  • Multi-Phase Batch Process Control. Batch processes were always difficult to analyse and control because there were so many variables changing at once and so many quite different sets of activities taking place in sequence. Users were not able to directly compare the many tens or hundreds of process variables and quality results across batches in a meaningful way let alone the reasons for the differences in performance of two or more apparently identical reactors. Process control has been limited to forcing the trajectory of a single variable, such as reactor temperature, to follow the same trajectory as that in a ‘golden’ batch. That was the past. Now they can compare many batches and many reactors with ease and go on to build realtime control models without maths.   

  • Operating Envelope Models for Continuous Process Control - A MIMO controller that requires no maths and no identification experiments and can work in open or closed-loop control above an existing regulatory control system. Includes RealTime Optimisation (RTO) too. Out-performed a leading MBC Controller in its first trial. An order of magnitude cheaper to implement and maintain than MBC too.

  • Improving Six Sigma and similar Continuous Improvement Methodologies - A 'No Equations' Six Sigma tutorial for Lean Manufacturing in process plants  (Six Sigma)

  • Process Data Mining without maths for the Support Specialists. Explore datasets containing hundreds of process and quality variables and tens or even hundreds of thousands of rows entirely visually using CVE's sophisticated but simple to use Query capabilities to drill deeper than you have ever drilled before. Discover black holes, different operating Modes, unsuspected relationships yet be able to communicate your findings easily to anyone on your site.

  • Quality by Design (QbD)

  • A new way to approach Design Of Experiments (DOE)

  • THE only complete solution for Process Analytical Technologies (PAT)

  • Quality Tracking

  • Formulation and scale-up

  • Production reporting

Case Studies let you see how GPC is being successfully implemented across different industries. You may well be experiencing the same problems as many of our customers were before they started using our software tools.

Curvaceous Software is the only business-to-business technology company specialised in the application of n-dimensional geometry to the business  problems of the process industries and in particular to operating cost reduction by decreasing operational variability in plants. Our technology is called Geometric Process Control (GPC) and is based on the parallel coordinate transformation.

Reducing operational variability will give you financial benefits in many ways such as: increased throughput, efficiency, quality, yield and extraction; decreased energy, recycle, waste, re-work, additives, chemicals, emissions and product inventory. The sum of all these will give you a substantial operating cost reduction.

You can use Geometric Process Control with or without an improvement management technology such as Total Quality Management (TQM) or Six Sigma, although our use of geometry improves the Six Sigma method and its effectiveness too.

Interactive visualisation provided by our products allows end-users to investigate, understand and solve operational problems without requiring an extensive mathematical background. This is in contrast to methods based on statistics, chemometrics, principal components analysis (PCA), partial least squares (PLS) or neural networks. The user of Geometric Process Control interacts directly with the full extent of his data instead of reducing it to a few artificial variables. Pre-formed hypotheses are not required, but can be used to advantage while existing opinions and beliefs about process operation are easily tested.

Geometric Process Control enables end-users to work much faster while gaining a better and more detailed understanding of how a process really operates. The speed of process improvement and process optimisation are greatly increased by the better understanding of the process conferred by the intuitive visual interface. Communication of the new understanding across an organisation is made much easier by visualisation.

Our Geometric Process Control products are in use in over 120 large process plants worldwide including Food, Pharmaceuticals (including primary or API manufacturing, secondary manufacturing, product formulation and product development), speciality and fine chemicals, commodity chemicals, cement, glass, paper, semi-conductors, pigments, minerals processing, polymers, petrochemicals, oil refining, offshore oil production, oil and gas exploration and drilling, defence, aerospace manufacturing. 

We provide our technology conveniently packaged in a small number of versatile and ready-to-use software products aimed at end-users performing functions such as Process Engineer, Production Engineer, Pharmacist, Process Chemist, Technician, Process Operator, Control Engineer, Improvement Engineer, Formulation Scientist.

 Operational Variability Reduction applications of Geometric Process Control for batch processes and continuous processes include

  • Everyday process problem solving or ‘fire-fighting’ to quickly answer questions such as “why isn’t my plant working as well as it was last week?”
     

  • Process Stewardship – daily monitoring by process engineers to ensure the plant works at its best all of the time and not just some of the time is made much faster and easier when the engineer can see and interact with several hundreds of variables at once in the parallel coordinates display. Continuous vigilance is needed to maintain an operating cost reduction.
     

  • Production Reporting

    • One parallel coordinate graph replaces many y-t graphs and shows overall achievement against multiple objectives over a period of time

    • GPC parallel coordinate charts are a multi-variable replacement for the SPC Charts used in statistical process control.
       

  • Finding better Operating Limits and Control Limits on many process and result variables simultaneously to increase achievement of a business objective. Takes much less time than existing methods.
     

  • Identifying Operating Modes of a continuous process is much easier when you can see them. There may be some you aren’t aware of because you couldn’t see them before. Now it is easy to give each Mode its own Operating Limits and Alarm Limits and even its own Operating Envelope. 
     

  • Batch Process Analysis to understand and reduce in-batch operational variability and run-to-run operational variability. You can compare many batches in one picture or compare the performance of several ‘identical’ batch reactors.
     

  • Consistent Alarm Limits allow better Alarm Management and a safer process. Geometric Process Control is the first-ever method to give numeric values for alarm limits for many variables simultaneously so that they are consistent alarm limits. They give earlier warning, giving the operator more time to react, with an immediate reduction in false alarms, standing alarms and alarm annunciation rate.
     

  • Shorter Alarm Rationalisation projects are the result of consistent alarm limit values being quickly and accurately calculated from an easily understood scientific basis which eliminates all the opinion and consequent time-consuming debate caused by present one-at-a-time methods. And you won’t need to repeat the alarm rationalisation project every five years as so often happens today.
     

  • Increased Throughput without sacrificing yield or quality can be achieved using GPC’s n-dimensional geometry Operating Envelope which brings further operational variability reduction by including and compensating for all variable interactions. The operating envelope is shown to the operator pictorially in terms of the existing process and result variables only with open or closed-loop guidance in how to keep the process inside the envelope and thus achieving the primary operating objective, which is usually the production of saleable product
    .

  • Intrinsically Safe Control. Operating advice provided by Geometric Process Control is intrinsically safe because of the underlying geometric basis of the envelope and advice algorithms. Only advice to move inwards into the known safe space of the Operating Envelope can be generated.
     

  • Inferential prediction of variables not measured in real-time improves process operation by eliminating sampling activity, waiting time and cost of laboratory analysis results. It is implicit in GPC’s Operating Envelope and requires no maths to implement for one or several variables. Prediction capability can be tested offline in GPC’s parallel coordinate chart.
     

  • Open-Loop control / Operator Guidance Control predominates for batch processes in the pharmaceutical and speciality chemical industries, possibly because there was previously no capable and affordable multi-variable closed loop controller available for batch processes. Open Loop control is provided by the GPC Operating Envelope with Closed Loop available by entering output destinations for control signals.
     

  • Closed-Loop control removes the operator from the loop increasing its reliability and ensuring that all control actions are implemented in a timely fashion. The GPC Operating Envelope includes cascade coupling and anti-windup capability.
     

  • Process Identification is not needed by GPC saving considerable project time and expense. A comparative trial in closed-loop control above a conventional PID control system gave better results than a commercial Model Based Control (MBC) system.
     

  • Minimum energy usage, minimum emissions, minimum cost, maximum throughput are typical Secondary Operating Objectives and are achieved in Geometric Process Control by finding sub-spaces within the Operating Envelope where both the primary and secondary objectives are achieved. This is Real-Time Optimisation. Objectives are specified by the Operator setting appropriate target ranges on process variables. No steady-state detector is required.
     

  • Generation and real-time display of Operator Alerts consistent with the previously-found High and Low Alarm Limits. The Alerts have the objective of preventing High and Low Alarms so further increase process safety by reducing the Probability of Failure on Demand (PFD) following a High/Low alarm annunciation.
     

  • Cycle time and product quality variability in complex batch processes can be reduced by using the Operating Envelope for Multi-Phase Batch Process Control. Sub-envelopes for each Phase are found without further end-user effort. Control can be open-loop operator guidance or closed loop.
     

  •  Fault Detection / Early Event Detection. Use of an Envelope as a means of detection has been found to give earlier detection of changed process behaviour than methods based on multi-variate statistics.
     

  • Time and experiments for Formulators and Scientists can be reduced through Response Surface Visualisation. The response surface is formed from the experimental values and results of even a small series of Designed Experiments (DOE) with no mathematical knowledge required. The experimenter or Product Formulator is given the capability using interactive visualisation to find for himself, without mathematics, the next experiment to perform. This saves him time and can result in fewer experiments.
     

  • Interactive Visualisation of the Design Space and Control Space defined by the FDA Quality by Design (QbD) initiatives leads directly to the Operating Envelope that is the unstated heart of the FDA Process Analytical Technology (PAT) initiative. Communication in visual form of the justification for the choice of a particular Design, Control or Operating space demonstrates the underlying process knowledge, and limitations of knowledge, that are at the heart of the FDA and ICH objectives in the registration of pharmaceutical products for human use.

 

   lean manufacturing. lean manufacturing. statistical process control. statistical process control. statistical process control.
 

 

 

 

 | Site Map | Legal Information | Copyright Curvaceous Software Ltd.  1998-2008  All Rights Reserved |