Deepa* et al.

International Journal For Research In Science & Advanced Technologies

ISSN: 2319-2690

Issue-1, Volume-1, 046-049

EXPERT SYSTEM FOR CAR TROUBLESHOOTING

Department of Computer Science, Shri Shankarlal Sundarbai Shasun Jain College for Women, [email protected], [email protected]

Abstract Due to advancement in technology getting a car

is cheap. But maintaining it is very expensive. It is also difficult to get a good mechanic; it will be helpful if we have some knowledge on automobile. We can rectify most of the problems. We can consult a mechanic if it is a major problem. Here we develop an expert system for car troubleshooting. The problems are broadly divided into two classes via car can be started with troubles and car brakes down. These major classes are further divided to identify the basic part, which creates the problem. The remedy is given if the problem is simple. If the problem is complex, the system will provide the causes. A. Index Terms Articifial Intelligence, Expert system, forward chaining, rule based.

is a decision charting tool in which the rules are simply defined by a combination of graphical shapes and pieces of text and is made available by logic programming associates. This can be of great use for people intending to purchase a property and also to legal experts for fast decision making. The design and development of a expert system aims to improve the method of selecting the best suitable faculty/major for student planning to be enrolled in.Al-Azhar university is discussed in[3].The basic idea of our approach is designing a model for testing and measuring the capabilities like intelligence understanding comprehension, mathematical concepts and others and applying the module results to a rule based expert system to determine the compatibility of these capabilities with the available majors. The development of appropriate knowledgerepresentations and natural language processing tools to provide the capability for an expert system to extend its own knowledge base by processing natural language text in machine readable form is crucial [4] describes the representation methods being used in the development of an early prototype for such a system. The knowledge is arranged in multilevel semantic network. The use of expert systems and artificial intelligence on particular the application of neural network to real estate forecasting is presented in [5].The use of Artificial Intelligence and Expert System have been suggested for a wider range of uses than is currently being research to any great extent. An important concept of the use of Expert System is that it is the overall system that is important rather than the reliance on new and better techniques. Interim results from a research project into knowledge representation facilitating explanation and Justification in knowledge based system is presented in [6].The research program has its roots in the re-development of an expert system Siratac, and the effect of this application on the acquisition and representation of the knowledge to facilitate explanations and knowledge Justification. An expert system for the evaluation of a dark fiber risk in the selling of wool, which explored the problem of providing suitable explanations of the domain knowledge for a naive end user, is discussed.

1.

INTRODUCTION

An expert System is a program that behaves like an expert in some, usually narrow, domain of application. Typical application include task such as medical diagnosis, locating equipment failures, or interesting measurement data. Expert systems have to be capable of solving problems that require expert knowledge in a particular domain. They should possess that knowledge in some form. Therefore they also called knowledge-based system. The occurrence of uncertainty statements with Artificial intelligence and expert system can be deal with probability [1].Probability is the only sensible description of uncertainty. The justification for the position rests on the formal, axiomatic argument that leads to the inevitability of probability as a theorem and also on the pragmatic verification that probability does work. A legal expert system for transfer of property act, a domain within the Indian Legal system which is often in demand is discussed in demand is discussed in [2].The Visirule software

[Jul-Aug

2012 Edition]

Page 46

Deepa* et al.
International Journal For Research In Science & Advanced Technologies

ISSN: 2319-2690

Issue-1, Volume-1, 046-049 The prototype represented the mapping between the various representations of the knowledge, or knowledge sources, as found in the expert system, namely the inference tree itself; research papers used as a knowledge source for parts of the inference tree. Artificial Intelligence research foundered on the issue of representation. When intelligence is approached in an incremental manner, with strict reliance on interfacing to the real world through perception and action, reliance on representation disappears. The approach to incrementally building complete intelligent creatures is discussed in [7].The fundamental decomposition of the intelligent system is not into independent information processing units which must interface with each other via representations. Instead, the intelligent system is decomposed into independent and parallel activity procedures which all interface directly to the world through perception and action, rather than interface to each other particularly much. Based on these principles we have built a very successful series of mobile robots which operate without supervision as creatures in standard office environments An approach to expert system explanation which is based on a classification of types of knowledge into reasoning domain knowledge, communication domain knowledge, and domain communication knowledge is discussed in[8].It is argued that this distinction, in addition to being theoretically appealing, allows us to better manage the software engineering aspect of explainable expert system development. The development of expert systems in medicine has generally been accompanied by a rejection of formal probabilistic methods for handling uncertainty is discussed in [9].We argue that a coherent probabilistic approach can, if carefully applied, meet many of the practical demands being made, and briefly illustrate our claim with three current projects. As we have illustrated with the chest disease system, updating of subjective probabilities is feasible and should provide a convergence of opinion that may overcome local biases which may otherwise render a system unacceptable. Historically, the study of artificial intelligence has emphasized symbolic rather than numerical computation. In recent years, however, the practical needs of expert systems have led to an interest in the use of numbers to encode partial confidence is discussed in [10].There has been some effort to square the use of these numbers with Bayesian probability ideas, but in most applications not all the inputs required by Bayesian probability analyses are available. This difficulty has led to widespread interest in belief functions, which use probability in a looser way. It must be recognized, however that even belief functions require more structure than is provided by pure production systems. The need for such structure in inherent in the nature of probability argument and cannot be evaded. Probability argument requires design as well as numerical inputs. The real challenge probability poses to artificial intelligence is to build systems that can design probability arguments. The real challenge artificial intelligence poses to statistics is to explain how statisticians design probability arguments. Research at the Dryden Flight Research Facility of the NASA Ames Research Center in the application of expert systems to problems in the Flight research environment is discussed in [11].In what is anticipated to be a broad research area a realtime expert system flight status monitor has been identified as the initial project. This real-time expert system flight status monitor is described in terms of concept, application, development, and schedule. In every car can built, the engine and chassis parts aresubjected to abrasive and corrosive wear, heat, vibrations and other factors that cause constant deterioration. In time, hard starting or failure to start, poor engine performance, low gas mileage, and other minor or more serious troubles may result from wear, broken parts or changes in adjustments. No other part of an automobile contributes more to safe operation than the brakes. Brakes must not only be able to stop a moving automobile, but they must stop it the shortest possible distance.Because the brakes are expected to decelerate an automobile at a faster rate than engine can accelerate it, the brakes must be able to control a greater amount of power than that developed by the engine. For this reason, the brakes on a modern high-powered automobile must be well designed and kept in top-notch operating condition by regular inspection, adjustment and repair.A cars troubl e symptoms may be simple and have an easily identified source, or they may complex. One can save time, wasted effort and money by using a planned procedure for troubleshooting that will. 1 .Quickly isolate the system responsible for the complaint. 2 .By elimination, determine the particular part or adjustment responsible for the trouble. An additional feature that is often required of an expert system is the ability to deal with uncertainty and incompleteness. Information about the problem to be solved can be incomplete or unreliable; relations in the problem domain can be approximate. A company can justify expert system development when the task solution has a very payoff. An expert system for mineral exploration could uncover a rich ore deposit worth millions of dollars. Expert system is justified when human experts are unavailable or unable to do the job. Often human experts are scarce, very much in demand and thus expensive. The problem is compounded when the company needs similar expertise at many different physical locations. Expert systems are justified when significant expertise is being lost to an organization through personnel changes. Retirement, job transfer often cause disruption and havoc because of the vital expertise that experienced personnel take

[Jul-Aug

2012 Edition]

Page 47

Deepa* et al.
International Journal For Research In Science & Advanced Technologies

ISSN: 2319-2690

Issue-1, Volume-1, 046-049 with them when they leave. The institutional memory aspects of an expert system minimize or even eliminate this problem. Expert system development is justified when the expert decision making must take place in an unfriendly or hostile environment. An expert system will perform well, that is, achieve the same levels of performance in the domain of interest that human experts can achieve. But simply producing good solutions is not enough. Real experts not only produce good solutions but often find them quickly. Thus an expert system must be skillful-apply its knowledge to produce solutions both efficiently and effectively, using the shortcuts or tricks the human experts use to eliminate wasteful or unnecessary calculations. When human experts solve the problems, particularly the type we consider appropriate for expert system work, they dont do it solving sets of equations or performing other laborious mathematical computations, instead they choose symbols to represent the problem concepts and apply strategies and heuristics to manipulate these concepts. An expert system also represents knowledge symbolically, as sets of symbols that stand for problem concepts. The consequence of this approach is that knowledge representation the choice, form, and interpretation of the symbols used becomes very important. An expert system has depth; that is it operates effectively in a narrow domain containing difficult, challenging problems. Thus the rules in an expert system are necessarily complicated, either through their individual complexity or their sheer number. Expert systems typically work in real world problem domains. An expert system has a knowledge that lets it reason about its own operation plus a structure that simplifies this reasoning process. For example, if an expert is organized as set rules, then it can easily look at the inference chains it produces to reach a conclusion. This knowledge the system has about how it reasons is called Meta knowledge, which means knowledge about knowledge. Modularity: each rule defines a small, relatively independent piece of knowledge. Incrementability: new rules can be added to the knowledge base relatively independently of other rules. Modifiability: old rules can be changed relatively independent of other rules. Support systems transparency. Rules are expressed as IF-THEN statements, as shown If the brakes pedal linkage is frozen Free up and lubricate linkage If undercharged battery Recharge battery These are rules that exist in the expert system for a cat troubleshooting. Rules are also written with arrows to indicate the IF and THEN portions of the rules. If the brakes pedal linkage is frozen lubricates linkage If undercharged battery free up and

recharge battery

(1)

Figure 1: inference chains

Car brakes down

Hard pedal

Frozen brake pedal linkage

II. METHODOLOGY
Rule based representations and forward chaining:

Rules provide a formal way of representing recommendations, directives or strategies, they are often appropriate when the domain knowledge results from empirical associations developed through years of experience solving problems in an area.In principle, any consistent formalism in which we can express knowledge about some problem domain can be considered for use in an expert system. IF-THEN rules turn out to be a natural form of expressing knowledge and have the following additional desirable features.

Free up and lubricate linkage An expert systems inference chains can be displayed to the user to help explain how the system reached is conclusions. In general, to develop a serious expert system for chosen domain we have to consult actual experts for that domain from experts and literature and modulating this understanding into chosen knowledge representation formalism is called knowledge elicitation. Forward chaining does not start with a hypothesis, but with some confirmed findings. Once we have observed that the hall is wet and the bathroom is dry, we conclude that there is a problem in kitchen; also, having noticed the kitchen window is closed, we infer that no water came from the outside; this leads to the final conclusion that there is a leak in the kitchen.

[Jul-Aug

2012 Edition]

Page 48

Deepa* et al.
International Journal For Research In Science & Advanced Technologies

ISSN: 2319-2690

Issue-1, Volume-1, 046-049 Forward chaining searches from data to goals, from findings to explanations or diagnoses etc. Since forward chaining starts with data we say that it is data driven. Reason for using forward chaining If there are many competing hypotheses, and there is no reason to start with one rather than another, it may be better to chain forward. In particular, forward chaining is more natural in monitoring tasks where the data are acquired continuously and the system has to detect whether an anomalous situation has risen; a change in the input data can be propagated in the forward chaining fashion to see whether this change indicates some fault in the monitored process or a change in the performance level. If there are a few data nodes and many goal nodes then forward chaining looks more appropriate EXPERIMENTAL RESULTS The expert system is developed in C language. CONCLUSION This paper presents an Expert system for car troubleshooting. We used rule based representation and forward chaining for identifying problems in the car and solution is given by the system REFERENCES [1] [1].Xu Zhao, Kai- 1.Dennis V.Lindley,The probability approach to the treatment of uncertainty in Artificial Intelligence and Expert system, Statistical Service,1987,vol 2, No 1,3-44. [2] 10.Glenn shafer,Probability judgment in artificial Intelligence and Expert systems,statistical Science,1987,Vol 1 [3] 11.Eugene L Deeke and Victoria A Regenie,Expert system Development and Application, National Aeronautics and space administration Oct 1985 [4] 2.Mr N.B Bilgi, Dr.R.V.Kulkarni.and Mr. Clive Spenser An Expert System using A Decision Logic Charting Approach for Indian Legal Domain With specific reference to Transfer of Property Act International Journal of artificial Intelligence and Expert Systems(IJAE),vol 1 Issue 2 [5] 3.Samy S.Abu Nasr, Mohammed H.Baraka and Abdurrahman Baraka, A Proposed expert system for guiding freshman students in selecting a major in AlAzhar University,Gaze,Journal of Theoretical and Applied information Technology,2005-2008 [6] 4.Julia E.Hodges,Jose L.Cordova,Lois M.Boggess, G.Jan Wilms and Rajee V AgarwalKnowledge base representation for an expert system capable of selfextension through natural language text analysis,Techinical Report MSU-900215,Dept of CS,Box 9637,Mississippi state,Ms 39762-9637. [7] 5.Peter Rossini,Using Expert system and Artificial Intelligence for real extate forecasting,Sixth Annual Pacific-Rim Real estate society conference,Sydney,Australia,24-27 January 2000. [8] 6.Bob Jansen,Two Expert Ssytem applications:inplications for knowledge representation for explanations and justifications,CSIRO Division of Information Technology,New south Wales,Australia. [9] 7.Rodney A Brooks,Intelligence without representation,MIT Artificial Intelligence Laboratory,Cambridge,USA. [10] 8.Regina Barzitay,Daryl Mecullough, owen Rambow,Jonathan Decristofaro,Tanye Korelsky and Benoit Lavoie, A new approach to expert system explanations,CoGenTex.Inc. [11] 9.David J.Spiegelhatter, Probabilistic expert systems medicine:pratical issues in handling uncertainity,Statistical science,1987,vol 2 ,No 1 3-44

[Jul-Aug

2012 Edition]

Page 49