| Financial analysis: |  |
| Environmental impact analysis: | --- |
| Waste management/P2: | |
| Environmental cost listing/database: | --- |
| Cost estimation: | --- |
| Alternative product/process comparison: | --- |
| Financial analysis: | |
| Environmental impact analysis: | --- |
| Waste management/P2: | |
| Environmental cost listing/database: | --- |
| Cost estimation: | --- |
| Alternative product/process comparison: | --- |
| Cost control | --- |
| Resource control | --- |
| Estimating control | --- |
| Schedule control | --- |
| Scope control | --- |
| Risk control | |
Another system, Neur-On-Line, is additionally offered in Alpha AXP systems. The system requirements are the same as for GDA. 48 MB of memory is required for use with GDA.
Real-time expert systems are on-line knowledge-based systems that combine analytical process models with conventional process control and heuristics to judge and interpret sensory data, while reasoning about the past, present, and future to assess on-going developments and plan appropriate actions. G2real-time systems can be applied to address the following environmental areas:
Real-time expert systems have been used proactively to minimize leaks, emissions, and waste, as well as to improve product quality, production yield, and energy consumption. They can assist in monitoring for regulatory compliance and identification and diagnosis of unexpected process changes. A real-time system can assist in waste minimization by using ingredient information for a compound to suggest production alternatives and methods to minimize wastes and production costs. It can also do mass-balance calculations.
The G2 Real Time Expert System is used to monitor complex industrial processes and to assist in problem identification. It combines rule and model-based expert system reasoning, real-time execution, user interface graphics, and monitoring system and database interface capabilities.
The G2 diagnostic assistant (GDA) can be used along with G2 and is a graphical, object-oriented decision-making tool. GDA contains predefined graphical blocks that represent information fed into the system through user- or computer-generated input. This information is used to interpret and diagnose real-time data received from controllers and sensors. GDA can be used for problem diagnosis and decision support. Users can also create their own applications within G2. Besides identifying problems after they have occurred, problems can also be predicted using statistical techniques. It can be used for prototyping, modeling, and simulation.
Neur-on-line is designed as a visual, object-oriented tool for building neural networks to monitor complex processes that are difficult to monitor or control with conventional modeling tools or rule-based systems. Neur-on-line has to be integrated with G2 (also GDA), which will perform tasks such as filtering input data and taking actions on the resulting outputs from the neuralnet.
There are on-line and off-line versions of the systems. A tool kit for modeling applications included in the packages allows users to modify the systems as required.
The Emission Reduction Research Center (ERRC), Massachusetts Institute of Technology (MIT), and Gensym are working on designing a Batch Design Kit using the G2 Real Time Expert System. This project was funded by a consortium of nine leading pharmaceutical and chemical companies, along with the US Army, New Jersey Commission on Science Technology, and the USEPA. The goal was to create a software toolkit for use in pharmaceutical and specialty chemical plants to reduce their environmental impacts (enable 'Zero Avoidable Pollution') by optimizing batch process design to minimize pollution. The Kit functions in a Computer Aided Design (CAD) environment for the design of batch processes with ecological constraints. Fed a list of ingredients for a given compound, the program will advise users of different production alternatives and help pinpoint methods that minimize wastes and production costs by selecting the most appropriate reactions; solvents, catalysts, diluents, etc.; reactors; separation and purification processes; and treatment units. General features of the Batch Design Kit include the following:
Idealized designs can be analyzed by enforcing environmental constraints, and more practical designs can be evolved by the guided relaxation of the constraints. Operating and capital costs can be allocated through integration with external databases. It is estimated that the kit would help companies save millions of dollars in non-compliance fines, waste product, and raw materials.
| Raw material acquisition | --- |
| Manufacturing stage | --- |
| Use/reuse/maintenance | --- |
| Recycle/waste management | --- |
Real-time systems can be used to address any part of the life cycle. Thusin real-time applications, it will typically cover stages within one facility that can be monitored. All life-cycle stages will not be addressed in one application. However, in off-line applications, the modeling capability could be used to include information on all life-cycle stages. For example, expert systems could be used in designing applications to help in selecting materials with optimal life cycles, while determining things like reusability of parts, ease of disassembly, best use of materials, etc. However, the system is not preset or designed for this purpose.
| Conventional | --- |
| Potentially hidden | --- |
| Contingent | --- |
| External | --- |
The system is not designed to include cost information, but users could possibly adapt the system to include some variable costs through relationship models. However, users would need to develop estimation methods, such as costs related to material use (and other real-time data),= or external costs related to emissions on their own. The systems can be used to control some cost drivers by maximizing efficiencies. For example, predicting future problems such as leaks or emissions could reduce contingent and external costs (without actually estimating or recording them). Real-time data will allow quicker identification and solution of problems and possibly less damage. In off-line versions, cost information could be incorporated in system models. Since the systems can be modified according to user requirements, any cost information, if available, could be incorporated. However, the system is not designed for estimation or resource control and therefore cannot assist in cost estimation.
The system is not designed for cost estimation.
| Net present value (NPV) | --- |
| Payback period | --- |
| Internal rate of return (IRR) | --- |
| Benefits cost ratio | --- |
| Other | --- |
Gensym's Telewindows application provides a multiuser client/server environment that allows multiple users to share G2 applications. The bridge products enable plant-wide integration through standard interfaces to popular database and control systems (PLC, DCS). User-defined blocks can be created in GDA. Users can build and maintain applications without needing to do conventional programming. G2 can import bitmapped images and create trend charts for displaying real-time data. G2's internal clock works at subsecond levels for scheduling G2 events and keeping data histories at subsecond time intervals. All systems have graphical user interfaces with the option to use MS Windows or MOTIF style interfaces. The new tool kit for modeling allows users to develop and modify the system application as required.
Gensym provides consulting services to its customers, including project prototyping, application engineering, project management, and customized training. Systems integration services include system architecture and data integration services. No demonstration disks are available. However, remote and on-site demos can be organized.
These systems are primarily designed to collect and process real-time data from automatic processes; therefore, they are unsuitable for small, nonautomated operations. The software is not designed as a cost-estimating system or to include cost data. It is meant primarily to be used to handle technical aspects of project planning and monitoring but not other non technical (resource control, administration, management) elements.
Return to the Table of Contents