More Technical Details

Our platform can interface with a variety of plant data sources including; databases, flat files, SCADA and DCS systems. Each individual data point or tag is sent to our application servers via the Internet using secure data transfer protocols. Each of our customers applications on the backbone is maintained in it own database. A typical database might include:

• Data Rate: 100 or more tags (sample points) at 1 sample per minute
• Short-term storage: several months of data available on line
• Long-term storage: several years of Min/Max/Average data
• Some installations may require more data to be available on-line, or a higher data rate.

The concept behind our platform is to provide a flexible framework for data visualization and analysis. The platform provides common services for use by all of applications. These common services include:

• Charting / Trending
• Reporting(automatically updated)
• Pattern Recognition
• Presentation (web pages)
• Common data transfer format (XML)

Providing users of iSagacity technology with assurance that their data is protected from unauthorized access is critical to the long-term success of our solution. There are a number of options for implementing security that not only insures the confidentiality of your data, but also impacts the performance and the ease of use of the platform by our customers.

Security
A detailed description of security issues can be obtained by contacting us. The following summarizes the key features of security on our platform:

• Data Transfer
  Data are sent from the client's site to iSagacity servers using standard Internet protocols such as ftp, udp, tcp, etc. As these protocols are not inherently secure, the data being transferred may be encrypted such that interception of the data stream will not yield any useful information. This can be accomplished using a public/private key pair exchange mechanism such as the Secure Sockets Layer (SSL) employed on consumer e-commerce sites if desired.
• Data Display
  Data displayed in HTML format over the Internet can be accessed by anyone, unless steps are taken to restrict access to particular pages. This is typically done by tying login/password information to user rights. In this way sites can be tailored not just to restrict viewing by unauthorized personnel, but also to tailor the data display to the particular user's position within the company.
• Authentication
  Authentication is the process of providing verification that users are who they say they are. It can be accomplished by obtaining a "Digital Certificate" from an issuing body that provides proof of identity. In a typical consumer e-commerce site, authentication is one-way. This means that the client machine can verify that the server is genuine but the server has no idea who the client is. This scheme works because the client's identity is verified when an order is placed by the credit card information supplied. In our solution, this one-way scheme may not be sufficient for very sensitive data. Although the data itself is encrypted and users are verified by username/password, it is still theoretically possible for a malicious user to pretend to represent a particular company and send fictitious data to the server. While a remote possibility, some users may consider this enough of a risk to require client-side certificates. In this case, each client machine would be issued a certificate (from iSagacity) that verifies identity. The downside of this approach is that in order for a user to access company data from the web, the machine they are using must have a certificate installed.
  Because there are different levels of security that may apply to different modules, a "one size fits all" approach is not appropriate. iSagacity customers can choose from a "cafeteria plan" that suits their particular needs.

Advanced Data Analysis
A key differentiator of our technology is the pattern recognition technology used to diagnose process data.  As described earlier, Remote Manager has the capability of being pre-configured with a library of patterns or data signatures. These data signatures can be developed through the use of a mathematical model or Simulator of your process. The model can be custom built from scratch using our engineering toolkit, or we can use a commercial Simulator from a third party. We can create a set of data signatures for your plant by Simulating faulted or abnormal conditions with the model to determine how the sensors in your system should respond to the modeled condition.

Process Data Miner develops the data signatures on the fly without operator interaction. There are several advantages to this approach. First, a model does not need to be developed. Model development can be costly and in many cases the development of representative models may not be possible. Second, Process Data Miner is continually learning new patterns in your data and organizing the data into a knowledge base. A pre-defined library of data signatures can only identify known causes and effects. Considerable knowledge can be gained by learning from your data directly. Third, Process Data Miner allows the data signatures to evolve over time without operator interaction. The performance of equipment and processes can slowly degrade that produce changes in patterns in your data. Models need to be continuously adjusted to account for this. Process Data Miner adjusts to this automatically.

Charting/Trending
A powerful feature of our platform is the ability to display user data on-line via the Internet in a variety of graphical formats. We allow users to view any plant data or output from our modules in a Simple to use web interface. The concept behind our technology is to provide all of its features in a “thin client” environment so the user only needs a standard browser. Please see anyone of the demos to see how this feature works. We also allow data to be easily downloaded into Excel at the click of a button. Many users find this to be a comfortable way of viewing and handling the data.

Reporting
A built in reporting tool is used to create custom web pages containing specific information typically extracted from user database(s) in a user-defined format. These tools permit us to custom design reports that convey specific information (i.e., weekly reports, etc.) for you that are of particular interest to plant personnel on a routine basis.