| j5 OMS |
|---|
|
Operations Management System j5 OMS is a broad range of hardened web applications that manage, control, organize & log the Operating Processes in industrial sites. Click here for more information... |
| j5 Logbook |
|---|
|
The pure Web Server version!! Make operator logs work for you! Industrial, proven, configurable. Oracle, SQL Server, Access web-server based. Click here for more information... |
| j5 HandoverBook |
|---|
|
The j5 HandoverBook is an electronic tool designed to effectively manage the information flow between outgoing and incoming shifts. Click here for more information... |
| What to look for |
|
|
What to demand from your Historian:A check-list.So you are it. You have been chosen to evaluate a historian. So many things to consider! (and where to start?) We put together a simple 18-point prioritized checklist of things to look for, things to avoid and what you should expect from the industry. Cost: If you can't justify the expense, then you are wasting your time and everyone else's too. Also, remember expensive systems are not necessarily any better. Industry norm? Aim for under one US$ per tag and approx $300 per user. (Depending on the number of tags in your system and the kind of database you are using.) Compatibility with your existing support network: If you have a team of Oracle experts in your company why go for an Access or SQL Server solution? (Or vice versa) Likewise, don't accept the "it's just a black box" pitch, it is not nice being in a position where your vendor knows more about your system than you do. Above all, these days, there is no reason why you should go for a proprietary database historian. Industry Norm? Insist on the database that you can support and make sure you can buy it from the database vendor not the historian vendor. Compatibility with your front end devices: If you are like 99.99% of the industries out there, then you will have more than one front end device on your factory floor. The last thing you are interested in is having to write low level drivers to link into a PLC, DCS or SCADA system or pay megabucks for someone to take an interminable time to write one for you. So avoid being caught by this "minor detail" which could end up costing you more than the rest of the project put together! Industry Norm? Insist that the historian supports OPC as a client so that you can go out and get interfaces for all your front end devices at a reasonable price. After all, this is exactly what OPC was designed to do! Simplicity and clarity: Difficult to define, but at the end of the day, exotic data structures that claim to be faster, more compact etc. will end up costing you big dollars to maintain and big dollars to implement. Industry Norm? Go for a simple structure and let the historian do the work not your engineers. The acid test? Ask the vendor to explain in one sentence how the data is structured in the tables! Visibility Toolbox: Once the history data is in the database, you are going to want to use it. (That is after all why you bought the historian!) You will need a range of tools from simple user-friendly viewing tools for large numbers of users to sophisticated query and development tools for the power user. Our toolbox would consist of (in order!):
Best Advice? Don't get wowed by the array of glitzy tools on the market. Instead, first carefully make a list of the tools you really need and will really use and only then make sure that they work with the historian you are evaluating. Configuration: You shouldn't have to type in tags and addresses to configure the system. Your historian should configure itself automatically and you should have facilities to filter out tags that you don't want to historize. Kind of data storage: You should expect your historian to be able to do various kinds of processing: For example, averages, maximums, minimums etc. You should be able to configure this processing and it should all be clearly visible in the table structure. Why would you want your engineers to write programs to do this? It is the job of the historian. The historian should also store a quality word for every data word stored. Scalability: Make sure that, as you expand, you are not going to run into any brick walls. Many of the items on the list below ultimately limit your scalability. Rate of data storage: Expect better than a million tags a minute on a modest Pentium III computer with say 256 MB of RAM. Number of users: Depending on your server and database, expect at least 25 and up to 100 or more simultaneous clients. Networked Historians: You should be able to add new historians into new areas and they should co-exist happily in a cluster arrangement.
|
