The probability of a human-caused incident occurring during dynamic positioning (DP) drilling operations is determined in this paper using binary logistic regression models built with data on 42 incidents that took place during the period 2011–2015. For each case, a range of variables characterising the configuration of the DP system, weather conditions and water depth are taken into account. These variables are taken into account to develop a logistic regression model that shows the likelihood of an incident being caused by human error. The results obtained show that human-based incidents are significantly more likely to occur when there is a lower usage of thrusters. These results are useful for focusing our attention on variables that may be associated with incidents attributable to human error, as well as for setting operational limits that could help to prevent these incidents and improve safety during these operations.

中文翻译：

---

确定动力定位钻井作业过程中人为失误造成事故的可能性

本文使用基于 2011 年至 2015 年期间发生的 42 起事故的数据构建的二元逻辑回归模型确定了在动态定位 (DP) 钻井作业期间发生人为事故的概率。对于每种情况，都考虑了表征 DP 系统配置、天气条件和水深的一系列变量。将这些变量考虑在内以开发逻辑回归模型，该模型显示事件由人为错误引起的可能性。获得的结果表明，当推进器的使用率较低时，人为事件更有可能发生。这些结果有助于将我们的注意力集中在可能与人为错误事件相关的变量上，