LACK OF PLANNING IN ALARM SYSTEM CONFIGURATION IS, IN ESSENCE, PLANNING TO FAIL

If the concept of alarming a process point is so simple, then why do s o many alarm system configurations end up as miserable failures? In ma ny cases, the causes can be traced back to the original configurations and the criteria used to select the points to alarm. Back when all al arms were hardwired (and the costs of alarming points had to be justif ied), a lot more thought was given to alarming points. Due to the deve lopment of computer based control and alarming, the costs associated w ith alarming a point have disappeared, along with much of the thought process of selecting alarm points. In many instances, alarms are added to a system with very little thought on how the alarm relates to othe r alarms in the system. The following paper is written based on experi ences in assessing and configuring alarm systems in the processing ind ustry over the past 14 years. Although much of the information for the paper was gleaned from experiences in the refining industry, many of the same problems appear in the batch processing industry. A good numb er of the problems happen during the initial configurations because a clear, concise alarming strategy is not set forth and followed during alarm system configuration. An alarm system need not be fancy nor empl oy cutting edge technology to be effective. The paper will outline ite ms that need to be considered prior to choosing points to alarm and as signing alarm priorities. Examples will be given of problems resulting from not setting forth and following an alarming strategy. (C) 1997 p ublished by Elsevier Science Ltd.