While advanced process control is the traditional way in which operators extract the last few fractions of a percent of efficiency out of a process, application of analytical engineering models to operating problems are complementary to that approach and solve a different set of problems as well as suggest different and sometimes more dramatic areas of improvement.
Employing an engineering model allows the plant engineer to investigate changes and modifications to a process which go beyond the simple tuning of existing process configuration as currently implemented.
This growing potential makes it even more critical to re-use the same models to solve different problems across the asset lifecycle and at different levels of granularity in operations. After all, a simulation that reliably predicts a particular application and situation becomes much more valuable if it can be applied to all tasks that require modelling of that unit or process. Indeed, the broader use of these models promises to have a profound business impact and  in most cases, the investment has already been made in building these models in the design of the plant. 
For the operator, the key opportunity is to ensure that this model, with its inbuilt understanding of the as-designed processes, is turned over to the operator. So, in this article, we'll describe current trends toward re-use of models and the integrated workflows that result.
First, let's set the stage by briefly summarizing the business challenges that are spurring the use of modelling technology to address a complete plant lifecycle

• Rapidly rising cost of energy and secondary cost of greenhouse gas emissions require the redesign and optimization of processes. 
• Global economic pressure imposes reduces investments in engineering, squeeze capital costs and force optimizing operations. 
• Shortages of skilled veteran engineers is continuing to grow more extreme. This demands increasingly powerful and easy-to-use models that capture organizational knowledge and experience.

These challenges call for moving to common models to solve multiple problems, making models simpler to use and integrating models with other software to solve broader business problems. Today's integrated modeling tools already attack many of these areas and the technology continues to evolve.

Key Trends in Modeling
Let's look at some ways in integrated modeling which are now providing value.
Simulation/economics work process 
The integration of economic analysis with the basic process development activity yields sizable benefits. Process costs are calculated and optimized concurrently with the conceptual process development, allowing the engineers to better understand the economic impact of their design decisions.
Fluor, which calls such integration 'cost optimized design', cites a number of benefits [1]. These include the ability to focus on technology/cost trade-offs early, improved quality of estimates and better cost awareness during design.
BASF estimates it achieves savings of 10 percent to 30 percent  in capital costs and up to USD2 million/year in energy through its Intelligent Total Cost Minimization (i-TCM) project approach, which involves performing process simulation, cost analysis and equipment modeling in parallel [2]. The goal is to optimize capacity, reduce operating costs and develop better designs for new or revamped plants.
Design/operability workflow The use of dynamic models for safety and operability analysis is another advance. This clarifies whether a design simulation solution is stable under real-world dynamic conditions. The goal is to use the same unit operations models for both steady-state and dynamic analysis, avoiding having to develop the models again.
Shell Chemicals takes this approach to model reactor and relief systems to ensure that designed safety systems will be able to contain any runaway reactions. This application of dynamic modeling improves operations safety and reliability and saves operating costs through optimized normal operations [3].
Conceptual/basic/detailed engineering workflow
Integrated basic engineering represents another area where workflows have advanced. The heat and material balance and flow sheets from simulation studies are directly input into the basic engineering process, where multiple disciplines define the FEED and then pass that information to detailed design.
WorleyParsons achieves an estimated 25percent increase in engineering efficiency and 50 percent reduction in time for basic engineering [4].
Moving models from R&D/engineering into plant operations Models developed during process development and design phases of a plant represent significant engineering effort and knowledge. Re-using those same models within the plant operating environment can provide even more benefits.
Process models suitable for use in plant operations span a spectrum from off-line steady-state simulation to debottlenecking analysis through to closed-loop real-time optimization of process performance. Table 1 highlights the different levels of benefit and implementation time and effort. Figure 2 illustrates the typical workflow in taking design models into operations.

Making the Transition
Off-line process models are the first step in re-using design models in operations. Because they serve an individual plant or operating unit, their topology is fixed and the range of operating conditions is well understood. The models are used for specific calculations, for purposes such as: advising on operating set points for individual equipment items; achieving a reconciled plant mass balance;  determining product properties; analyzing energy usage,  comparing actual versus design performance; responding to changing market conditions;  meeting product specifications and  retaining and enhancing process knowledge.
Even though they may connect to real-time data systems, off-line models are not fully automated, a person normally initiates runs.
Models produced during the design phase usually require additional work before they can be used as off-line process models. If difficulties such as convergence failure occur, the design engineer who understands the constraints of the model and the range of valid conditions may need to advise.
For use in operations, the model must be tuned to match plant conditions and the particular calculations being executed. For instance, the plant set-up may change from day to day  with different product grades being produced and individual units or controllers switched on/off. The off-line process models must account for these specifics.
In addition simulations only are valid within a limited range of operating conditions that must be understood and enforced. Models will need to be made robust, to ensure they converge within the valid operating ranges. Model inputs, both those entered manually and those coming from real-time data systems, must be kept within these ranges, this often is done by running the models not through their normal 'engineering' user interface but through a simpler custom interface, (such as one based on Excel) to ensure they are run properly. The sidebar provides other practical pointers for moving models from design to operations.

The Next Steps
If an off-line process model gets regular use in operations, it may be appropriate to convert it to a real-time open-loop model. The model execution then can be automated to occur, say, once per shift or every N minutes or when triggered by a process event. Such open-loop models also may write results back to the plant's real-time data systems. However, the results of the model always are evaluated by a person, who ultimately accepts or rejects any advice or data.
Additional effort is necessary to make these automated models even more robust, read in additional real-time plant data and reconcile conflicting plant data (eg, measured mass flows in and out of a unit that don't balance).
The final level of modeling in operations employs real-time closed-loop models, with their results implemented in an automated way to optimize processes. These systems require additional effort to make the system fully robust and safe. However, they promise even greater benefits, particularly where processes need to respond to predictable variability (eg, in feedstock characteristics).
Chemical companies already are realizing significant benefits in plant operations from each of these approaches [5]. The experience reported by INEOS is instructive. It used a modeling approach to optimize heat exchanger monitoring and cleaning in its vacuum distillation units, saving over $3 million dollars per unit per year [6].
 
Future Directions
Innovation in integrated engineering continues along two fronts: for collaboration between engineering disciplines and for moving rigorous analytical models into the operating environment. Here are some key developments
Modularized systems Process modeling systems can be redesigned for re-use in a modular fashion throughout an asset's lifecycle. One example is the physical properties database. AspenTech now offers its as a re-usable resource, a 'standardized component' for a number of different model-based applications. This ensures maximum flexibility and consistency regardless of choice of modeling tools.
Another example is the unit operations models. These can be modularized so they are usable by systems ranging from simulation, basic engineering, optimization, economic evaluation to advanced process control.
User console and simplicity New concepts build the workflow right into the user interface presenting the appropriate analytical models and tools to users depending upon their role, the phase of a project and their position in the workflow.
Models in engineering This provides the capability to call models from downstream in the design process, including basic design, start-up and control without looping back to the modeling group. Modeling can be performed in-plant without intervention by design engineering.
Common engineering data backbone A lifecycle database incorporates unit operations models, process, equipment and instrumentation data and control information to facilitate lifecycle optimization.

Realize Real Benefits
Process engineering models created during conceptual design increasingly are being applied downstream in the design process and operations, thanks to developments that make these analytical models usable by other disciplines and plant staff. This is leading to measurable savings in dollars, energy, time and staffing.
For example, AspenTech, employing engineering models on behalf of a Mediterranean based petrochemical company made an operating strategy modification involving feed to a distillation column that resulted in several million dollars per year energy savings.  This operational improvement was not discovered by conventional APC or plant performance analysis techniques.

Ron Beck is the Marketing Manager for Basic Engineering with Aspen Technology in Burlington, Mass.
Email: ron.beck@aspentech.com.
Rob Hockley is a  U K -based senior consultant for Aspen Technology, Inc and  can be contacted
Email: rob.hockley@aspentech.com