This project aims at developing novel approaches and techniques for modern Process-Aware Information Systems (PAIS) by integrating Business Process Management (BPM) and IoT Big Data technologies. Enabled by IoT, BPM can perceive real-time physical world events, which results in faster and better decisions reacting to them. At the same, the ever-increasing IoT big data can be effectively leveraged and programmed by means of business processes. Focusing on the scientific research challenge of paradigm misalignment between BPM and IoT, this project plans to deal with three fundamental issues, namely, mismatch of programm-ing mechanisms, mismatch of data persistence mechanisms, and mismatch of ad hoc processes and overall control, to make breakthroughs in the abstraction of IoT data and events, proactive programming, and instantaneous process adaptability induced by IoT events, so as to make traditional BPM more proactive. A series of test cases and a set of benchmarks will also be validated based on the typical application scenarios from Adesso SE in Germany. The main innovative contributions include:
       1) an enriched BPM modeling approach with novel abstracting and programming mechanisms to facilitate the induction of IoT big data, modeling of proactive data services, and an IoT-enabled micro-process model to realize situation-aware IoT big data processing logic;
       2) the topological models for IoT-enabled micro-processes, their organization and evolution mechanisms. To this end, a massive amount of ad hoc micro-processes will be organized and managed in a controllable manner;
       3) a data persistence model that reconciles stream data and batch data, which contributes to unified management and query in decentralized settings;
       4) an agile solution to long-tail variants of business processes that cannot be properly handled by traditional process adaptation, and the controllability verification against some key performance indicators (KPIs). All the above achievements will support proactive service routing and evolutionary learning driven by IoT-related unpredictable events and situations.

       本项目旨在促进物联大数据（IoT Big Data）和业务过程管理（BPM）的融合，BPM借助IoT可更快更准地感知物理世界，并为不断膨胀的物联大数据提供应用编程范型。针对范式适配这一个关键科学问题，突破三大难点—编程机制失配、数据模型失配、临机流程与全局控制失配，取得抽象和编程机制及应变方法上的创新，变传统BPM为主动式BPM，并基于国际大型IT服务提供商adesso SE实际场景建立验证靶场和测试基准。主要产出有：
       1)适于物联大数据的服务抽象和编程方法，提出IoT微流程模型和主动式BPM算子，快速便捷地集成物联大数据驱动的临机逻辑;
       2)微流程的复杂网络拓扑模型及其演化与发现机理，可有序组织和利用大量IoT临机协作片段;
       3)流批融合的一体化数据模型，实现去中心环境下IoT和BPM数据的一体化管理和查询;
       4)IoT使能的BPM应变方法与合规的业务目标可控性判定理论，应对过程模型长尾型变化。