Resilience engineering (RE) has introduced new ideas into reliability engineering and safety science and has highlighted new concepts of safe work that could be considered for reliability improvement. The aim of this paper is to put forward an approach within the framework of resilience engineering to service reliability improvement in maintenance organizations. The strategy for reliability improvement is based on mutual learning through collaboration among maintenance departments of a public gas company in 11 districts. The proposed approach first assesses RE in the maintenance departments using a standardized questionnaire. Using a linear regression model, then the relationship between RE and service reliability in the maintenance departments is identified. Service reliability is defined as the probability of undesirable events: non-working valves, gas leaks, unavailability of valves due to coverage by asphalt, and gas flow cut-offs. The pros and cons of departmental collaborative learning in terms of absorbing the RE best practices, managerial or political conflicts, and collaboration costs are discussed. Finally, a linear integer programming is formulated to find the best collaborative connections so that the maintenance departments could find those peers from whom they can benefit most, boost their RE practice, and finally improve their service reliability, while minimizing their costs and possible conflicts.

弹性工程（RE）已将新思想引入了可靠性工程和安全科学领域，并着重强调了可以考虑提高可靠性的安全工作的新概念。本文的目的是在弹性工程学框架内提出一种方法，以提高维护组织的服务可靠性。可靠性改进策略基于11个地区的一家公共天然气公司维护部门之间的相互学习。拟议的方法首先使用标准化的调查表评估维护部门的可再生能源。使用线性回归模型，然后确定维护部门中RE与服务可靠性之间的关系。服务可靠性定义为不良事件的概率：阀门不工作，气体泄漏，由于沥青覆盖而导致的阀门不可用以及气流截止。讨论了部门协作学习在吸收可再生能源最佳实践，管理或政治冲突以及协作成本方面的利弊。最后，制定了一个线性整数规划程序，以查找最佳的协作连接，以便维护部门可以找到他们可以从中受益最大的对等点，提高他们的RE实践，并最终提高他们的服务可靠性，同时最大程度地降低成本和可能的冲突。