Application Of Expert Systems In Offshore Structure Installation

Abstract

During the installation of various components of offshore structures, two of the important tools the construction vessel superintendent uses are her experience and judgement. These however cannot be expressed in rigid mathematical rules and expressions. Thus, in the past the installation of offshore structures has been somewhat people specific. Of late, a branch of computer science namely artificial intelligence and especially one of the areas it supports can be used to formula these rules. This emerging area of computer science is called expert systems and these rules are known as knowledge based production rules. These are heuristic rules that can be manipulated to draw an inference. The paper describes the concept of knowledge based expert systems and its' application in solving installation problems of offshore structures. Incidentally, all personal pronouns in this paper are feminine. The author requests that whenever you read the words 'she' and 'her' you take it to mean 'he' and 'his' as Well. INTRODUCTION: It is possible today through the simulation of sophisticated algorithmic computer programs to decide on certain tangible aspects of offshore operations relating to the installation of structures. For example, we are able to accurately predict the ballasting sequence for the upending of a jacket or the laying profile of a particular pipeline. Mathematical rules or expressions dictate these predictions. However, there are other intangible aspects of installation of offshore structures which so far have been performed by human expertise. A few examples of these are how do you decide to suspend pipelaying operation or start transferring pipelines from material barge to lay barge or how do you determine if the weather is good for launching a jacket? The construction vessel superintendent learns her trade through years of experience. She then uses it to perform her work. The rules that control her decisions are not dictated by any mathematical rules and therefore can not be predicted by any algorithmic computer program. These are heuristic rules and have traditionally been treated as enigmas. And this situation has remained an acceptable status quo. There are some problems however. Human expertise is perishable, difficult to transfer and document and, could be unpredictable. It could also lead to heterogeneous offshore operation. If a superintendent leaves a corporation, the 'intangible' knowledge goes with her. A branch of computer science called artificial intelligence deals with artificial expertise. The subject of expert systems is a practical application of artificial intelligence. The expert systems can manipulate a set of heuristic rules and draw conclusions from it. From reference #1, "Expert systems are a class of computer programs that can advise, analyze, categorize, communicate, consult, design, diagnose, explain, explore, forecast, form concepts, identify, interpret, justify, learn, manage, monitor, plan, present, retrieve, schedule, test and tutor. They address problems normally thought to require human specialists for their solution." This definition certainly can not be criticized as unduly restrictive. Expert systems seem to be a likely candidate for usage in installation of offshore structures. Artificial expertise has certain advantages over human expertise.