When a given water distribution network is proficiently installed, the most likely hitch that could result in limiting the discharge of its function is failure of any of the facilities. Pipes have suffered the worst heat in campus communities and hence, deserve unending attention. The selection of enabling maintenance planning and control technique for specific water network scenario and infrastructural (pipe) conditions is cardinal in the availability and sustenance of efficient water distribution. This paper explores the use of artificial neural network to analyse and manage water pipe failure in a Campus community. An illustrative example has been demonstrated with the dominant AMC pipe failure dataset obtained from a typical Campus community, namely; University of Nigeria Nsukka. The indices of performance employed in the model include the mean absolute error which was 0.004052 and coefficient of determination (0.99548) which represents a very good fit. Deterioration curves were used to elicit the relationship of the failure variables on failure span. The results show that there is a strong correlation between the pipeline failure variables with the failure span. The average pressure head was closely directly proportional to the time of next failure while the number of previous pipe failures is inversely proportional to the time of next failure. This revelation is an important milestone which goes to supplement the decision tools of the maintenance practitioners in the studied location and the likes.