Abstract: To solve the problem of mismatch between role allocation,authority and ability in business process collaboration,a role collaborative tuning algorithm by combining professional ability evaluation model and role authority was proposed.Based on the business process log,a time series prediction model was constructed to detect the inefficient process instances.Moreover,a loop mining algorithm named Meta Path-based Loop Mining Algorithm (MPLMA) was constructed to mining roles collaboration pattern.Based on the business process log,a professional ability evaluation model was built to compute the roles ability.Combining professional competence evaluation model with authority judgment method,the role collaboration mode was optimized.The experimental results showed that the proposed MPLMA could effectively mine collaborative instances,and the proposed role collaborative optimization algorithm could accurately and reasonably guide the optimization of collaborative patterns.

Key words: business process, time series, meta path, role authority, disposable resources

摘要： 为解决业务流程协同模式中存在的角色能力、权限与职责不匹配的问题,综合考虑角色能力与角色权限,设计了基于角色专业能力评价模型的协同调优算法。首先基于业务流程的数据库操作日志,建立了时间序列模型检测异常业务流程,提出基于元路径的环路挖掘算法(MPLMA),从而挖掘业务流程中由协同导致的重复工作;以数据库的操作和流程日志为基础,建立专业能力评价模型,量化评价角色能力;将专业能力评价模型与权限判定方法相结合,优化角色协同模式。实验结果表明,所提基于元路径的环路挖掘算法能够高效率地挖掘出协同实例,所提出的角色协同调优算法能够精确、合理地指导协同模式的优化。

关键词: 业务流程, 时间序列, 元路径, 角色权限, 可支配资源