News Item: Federal agents are searching for a shipment of radioactive material that has been unaccounted for since it crossed the US/Canadian border at either Port Huron or Detroit in May. Sensors to detect material at the border showed positive readings when checked days after the truck apparently passed, the Detroit Free Press reported August 3. Government inspectors first became aware of the missing shipment in early June--about a week after it crossed the border, officials said.

Situation Analysis: The problem of the missing radioactive shipment is a frightening example of an all-too-common problem in the design of information supply chains in businesses--the organization invests heavily to create near-real-time data capture and analysis, only to discover that it does not benefit from that investment because it has failed to change its business processes to absorb and act on the data in real time.

In this case, the border sensors captured evidence that a radioactive shipment was crossing the border in real time, but because the business processes to recognize and act on that information (e.g., to raise an automated gate or other device to stop the truck) were not in place, the shipment remained unidentified. The likelihood is that it was a completely innocent shipment (e.g., material for medical purposes), but the possibilities are frightening.

[an error occurred while processing this directive]

"The typical information supply chain has 20 links," says Meta Group analyst Doug Laney. "And it is only as strong as the weakest of those links."

Organizations looking at moving from batch to real-time business processes need to consider several issues, the first of which is what is meant by "real time."

"Actual real-time information supply chains are very expensive to build and maintain," warns Meta Group analyst Aaron Zornes. "Therefore, actual real time is confined to specialized uses in the financial services industry and the military command-and-control infrastructure, where any delay can be costly."

The rest of the world is increasingly focusing on near-real-time response, which can mean anything from a few minutes to an hour or more depending on the time value of the information.

Even this, however, requires a large investment. Many business processes do not lend themselves to near-real-time operations--they are batch by nature. Others are simply not worth the investment of moving from batch to real time.

"Organizations should not fall into the trap of trying to redesign every process to move to real time all at once," warns Laney. "They should not jump on the real-time bandwagon with both feet and create an information nightmare." Instead, companies need to understand and concentrate on the processes that will provide the greatest impact, while leaving the vast majority of data collection, analysis, and business process in batch mode. Further, the information supply chains that are selected must be completely redesigned.

"Real-time analytics does not mean real-time business performance," says Laney. "An executive dashboard can be updated from the data warehouse every minute, but if the executive is not watching the dashboard or does not understand the business implications of the changes it shows--or if the changes do not have important business implications--that does no good."

1 2	| Next page

Newsletters
	Tech Update Today eBusiness Update	Tech Update Weekly		All newsletters FAQ Manage my newsletters

[an error occurred while processing this directive]

[an error occurred while processing this directive]