Ubiquitous Computing is considered as enabler for linking everyday life with information and communication technology. However, developing mobile and pervasive applications that provide personalized and flexible user assistance still constitutes a challenge. Relevant scenarios are diverse and encompass domains like healthcare, logistics, and business collaboration.
On the one hand, daily business activities like customer services or long term medical care are often executed in a mobile manner. On the other hand, IT process support for them usually stops at workstation level. Consequently, information gathered by mobile workers or mobile users is entered with delay, which entails problems like falsification, loss of information, delayed access, and lack of control. From the integration of mobile devices, we expect a new quality in human-centric process support.
Research focus of MARPLE is to enable a tight integration of process management technology and mobile computing. Our overall vision is to provide sophisticated mobile process support; i.e., to realize generic process management features on mobile devices including support for process changes, process instance migrations, process enactment, and process monitoring. We base our work on core design principles and fundamental concepts we developed in our ADEPT project (e.g., correctness-by-construction, dynamic process adaptations, etc.).

Fig. 1. depicts an simplified use-case from the healthcare domain. When adding mobile devices to environments like shown in Fig. 1. (with running processes and partly executing those processes on these devices) challenging issues emerge. As indicated in Fig. 1, the overall process is fragmented and some of these fragments are deployed to mobile devices. In particular, the following challenges exist in this context:

• It must be possible to partition a process model and to allocate the resulting fragments on mobile devices as well as on personal workstations.
• Soundness of the overall process (i.e., the process choreography) needs to be ensured, also in the context of distributed process instance changes.
• The runtime infrastructure has to cope with problems like broken connections or malfunctioning devices.
• When running the fragments on distributed devices their execution must be efficiently synchronized and messages be exchanged in a reliable way.
• Both the overall process model as well as its fragments may have to be adapted during runtime (e.g., to deal with exceptional situations).
• A mobile process must be able to gather sensoric data during its execution as well as to include additional context information (e.g., location of mobile workers).
• The whole life cycle of process management and mobile applications needs to be covered.

To foster our vision, we are working on a prototype enabling mobile assistance. Its kernel comprises a light-weight process engine that runs on the mobile device and that is able to interact with backend processes if required. In addition, we provide advanced tools for configuring and verifying process fragments. When developing our MARPLE engine, we had one shining example in mind - the ADEPT process management system we had developed during the last decade. In particular, we adopt basic design principles from ADEPT (e.g., correctness-by-construction, dynamic process adaptability), but additionally align the MARPLE architecture with specific needs of mobile processes. The MARPLE architecture is depicted in Fig. 2.

Our MARPLE architecture consists of two main components: Mediation Center and Mobile Engine, (cf. Fig. 2). The former runs on a workstation and enables the modeling, configuration, deployment, and control of mobile processes. The Mobile Engine includes mechanism to migrate running processes from one device to another, as well as to support ad-hoc changes of processes running on the mobile device. In addition, the Mediation Center governs accrued entities (e.g. process templates, instance data, and device configurations).