An age-based replacement model for repairable systems with attention to high-speed marine diesel engines

Abstract

The background of this thesis is the high number of severe failures of high-speed propulsion engines in high-speed vessels. When number of failures is high, the preventive maintenance comes into particular focus. Hence, this study will consider the interval (age), at which to carry out scheduled engine replacement or renewal by overhaul.

The main objective shall be to develop a new age-based replacement model for repairable systems that combines system reliability theory with observed system failures and maintenance costs in order to determine the optimal age for performing scheduled replacement. The presented model classifies failures into either minor or major. Minor failures are minimally repaired while major ones initiate unscheduled replacement. A maintenance cost model for determining the optimal age for scheduled replacement is established.

Compared to the basic age-replacement model, the new one is more realistic, by including cost of both minor and major failures, which should produce a more costeffective optimal age for scheduled replacement. The model is designed to provide decision support for the management on when to replace or overhaul. It can represent a valuable correction to the general recommendations given by the manufacturer.

Despite the model background of high-speed marine diesel engines, it should be suitable also for other repairable systems which are subject to a replacement policy.

The applicability of the model is demonstrated by estimating the optimal age (interval) for performing scheduled major overhaul of a high-speed propulsion engine in a fast passenger ferry. The findings indicate that practiced scheduled age at major overhaul may be somewhat too high. The findings also indicate that the manufacturer’s general recommendation regarding age at major overhaul is far from optimal for the engines in the ferries referred to in this case study.

Hopefully, this thesis will also contribute to a better understanding of the value of recording maintenance and failure events and how these data can assist the maintenance engineers and management in making better decisions.