Processes tend to interact with other processes and operate on various objects of different types. These objects can influence each other creating dependencies between sub-processes. Analyzing the conformance of such complex processes challenges traditional conformance-checking approaches because they assume a single-case identifier for a process. To create a single-case identifier one has to flatten complex processes. This leads to information loss when separating the processes that interact on some objects. This paper introduces an alignment approach that operates directly on these object-centric processes. We introduce alignments that can give behavior-based insights into how closely related the event data generated by a process and the behavior specified by an object-centric Petri net are. The contributions of this paper include a definition for object-centric alignments, an algorithm to compute them, a publicly available implementation, and a qualitative and quantitative evaluation. The qualitative evaluation shows that object-centric alignments can give better insights into object-centric processes because they correctly consider inter-object dependencies. Findings from the quantitative evaluation show that the run-time grows exponentially with the number of objects, the length of the process execution, and the cost of the alignment. The evaluation results motivate future research to improve the run-time and make object-centric alignments more applicable for larger processes.

翻译：流程往往与其他流程交互，并操作不同类型对象。这些对象可能相互影响，从而在子流程之间形成依赖关系。分析此类复杂流程的一致性对传统一致性检查方法构成挑战，因其假设流程存在单一案例标识符。为生成单一案例标识符，必须对复杂流程进行扁平化处理。这会在分离交互对象的相关流程时导致信息丢失。本文提出了一种直接作用于这些以对象为中心流程的对齐方法。我们引入了一种对齐机制，能够从行为层面揭示流程生成的事件数据与以对象为中心的Petri网所规约行为之间的关联程度。本文的贡献包括：定义以对象为中心的对齐、提出计算算法、公开实现代码以及定性与定量评估。定性评估表明，以对象为中心的对齐能更深入洞察以对象为中心的流程，因其正确考虑了对象间依赖关系。定量评估结果显示，运行时间随对象数量、流程执行长度及对齐成本呈指数增长。该评估结果激励未来研究改进运行效率，使以对象为中心的对齐更适用于大规模流程。