Yutian Sun
University of California, Santa Barbara
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Yutian Sun.
international conference on service oriented computing | 2013
Fenno F. Terry Heath; David Boaz; Manmohan Gupta; Roman Vaculín; Yutian Sun; Richard Hull; Lior Limonad
A promising approach to managing business operations is based on business artifacts, a.k.a. business entities with lifecycles [8, 6]. These are key conceptual entities that are central to guiding the operations of a business, and whose content changes as they move through those operations. A business artifact type is modeled using a an information model, which is intended to hold all business-relevant data about entities of this type, and b a lifecycle model, which is intended to hold the possible ways that an entity of this type might progress through the business. In 2010 a declarative style of business artifact lifecycles, called Guard-Stage-Milestone GSM, was introduced [4, 5]. GSM has since been adopted [7] to form the conceptual basis of the OMG Case Management Model and Notation CMMN standard [1]. The Barcelona component of the recently open-sourced [2] ArtiFact system supports both design-time and run-time environments for GSM. Both of these will be illustrated in the proposed demo.
international conference on service oriented computing | 2012
Yutian Sun; Wei Xu; Jianwen Su
A choreography models a collaboration among multiple participants. Existing choreography specification languages focus mostly on message sequences and are weak in modeling data shared by participants and used in sequence constraints. They also assume a fixed number of participants and make no distinction between participant type and participant instances. Artifact-centric business process models give equal considerations on modeling both data and control flow of activities. These models provide a solid foundation for choreography specification. This paper makes two contributions. First, we develop a choreography language with four new features: (1) Each participant type is an artifact schema with (a part of) its information model visible to choreography specification. (2) Participant instance level correlations are supported and cardinality constraints on such correlations can be explicitly defined. (3) Messages have data models, both message data and artifact data can be used in specifying choreography constraints. (4) The language is declarative based on a mixture of first order logic and a set of binary operators from DecSerFlow. Second, we develop a realization mechanism and show that a subclass of the choreography specified in our language can always be realized. The mechanism consists of a coordinator running with each artifact instance and a message protocol among participants.
international conference on data engineering | 2014
Yutian Sun; Jianwen Su; Budan Wu; Jian Yang
An important omission in current development practice for business process (or workflow) management systems is modeling of data & access for a business process, including relationship of the process data and the persistent data in the underlying enterprise database(s). This paper develops and studies a new approach to modeling data for business processes: representing data used by a process as a hierarchically structured business entity with (i) keys, local keys, and update constraints, and (ii) a set of data mapping rules defining exact correspondence between entity data values and values in the enterprise database. This paper makes the following technical contributions: (1) A data mapping language is formulated based on path expressions, and shown to coincide with a subclass of the schema mapping language Clio. (2) Two new notions are formulated: Updatability allows each update on a business entity (or database) to be translated to updates on the database (or resp. business entity), a fundamental requirement for process implementation. Isolation reflects that updates by one process execution do not alter data used by another running process. The property provides an important clue in process design. (3) Decision algorithms for updatability and isolation are presented, and they can be easily adapted for data mappings expressed in the subclass of Clio.
business process management | 2013
Rik Eshuis; Richard Hull; Yutian Sun; Roman Vaculín
Case Management is emerging as an important paradigm for Business Process Management. The Guard-Stage-Milestone (GSM) model is a recent case management approach that substantially influences OMGs emerging Case Management Modeling Notation standard. We study the problem of outsourcing part of a GSM schema to another party, and develop a framework that supports splitting and outsourcing of GSM schemas. One element of the framework focuses on restructuring the GSM schema to facilitate outsourcing while preserving the semantics of the original schema; the second focuses on locking protocols that define how the distributed parties should operate. Additionally, the framework allows parties to keep local parts of their GSM subschema private without affecting the outcomes of the global execution. The rules restructuring developed here enables a crisp separation of concerns, which allows reuse of existing GSM (and thus Case Management) engines for executing the subschemas.
business process management | 2014
Yutian Sun; Jianwen Su; Jian Yang
In most business process management (BPM) systems, the interleaving nature of data management and business process (BP) execution makes it hard for providing “Business-Process-as-a-Service” (BPaaS) due to the enormous effort required on maintaining both the engines as well as the data for the clients. In this paper we formulate a concept of a self-guided artifact, which extends artifact-centric BP models by capturing all needed data for a BP throughout its execution. Taking advantage of self-guided artifacts, the SeGA framework is presented to support the separation of data and BP execution.
acm transactions on management information systems | 2016
Yutian Sun; Jianwen Su; Jian Yang
In most BPM systems (a.k.a. workflow systems), the data for process execution is scattered across databases for enterprise, auxiliary local data stores within the BPM systems, and even file systems (e.g., specification of process models). The interleaving nature of data management and BP execution and the lack of a coherent conceptual data model for all data needed for execution make it hard for (1) providing Business-Process-as-a-Service (BPaaS) and (2) effective support for collaboration between business processes. The primary reason is that an enormous effort is required for maintaining both the engines and the data for the client applications. In particular, different modeling languages and different BPM systems make process interoperation one of the toughest challenges. In this article, we formulate a concept of a “universal artifact,” which extends artifact-centric models by capturing all needed data for a process instance throughout its execution. A framework called SeGA based on universal artifacts is developed to support separation of data and BP execution, a key principle for BPM systems. We demonstrate in this article that SeGA is versatile enough to fully facilitate not only executions of individual processes (to support BPaaS) but also various collaboration models. Moreover, SeGA reduces the complexity in runtime management including runtime querying, constraints enforcement, and dynamic modification upon collaboration across possibly different BPM systems.
OTM Confederated International Conferences "On the Move to Meaningful Internet Systems" | 2012
Yutian Sun; Richard Hull; Roman Vaculín
The business artifact (a.k.a. business entity) approach to modeling and implementing business operations and processes is based on a holistic marriage of data and process and enables a factoring of business operations based on key business-relevant conceptual entities. The recently introduced Guard-Stage- Milestone (GSM) artifact meta-model provides a hierarchical and declarative basis for specifying artifact lifecycles, and is substantially influencing OMG’s emerging Case Management Modeling Notation standard. In previous papers one characterization of the operational semantics for GSM is based on the incremental, strictly serial firing of Event-Condition-Action (ECA) like rules. This paper develops a parallel algorithm equivalent to the sequential one in terms of externally observable characteristics. Optimizations and analysis for the parallel algorithm are discussed. This paper also introduces a simplification of the GSM meta-model that provides more flexibility and makes checking for well-formedness of GSM models simpler and more intuitive than in the preceding works on GSM.
Information Systems | 2014
Rik Eshuis; Richard Hull; Yutian Sun; Roman Vaculín
Case Management is emerging as an important paradigm for Business Process Management. The Guard-Stage-Milestone (GSM) model is a recent case management approach that substantially influences OMG@?s emerging Case Management Modeling Notation standard. We study the problem of outsourcing part of a GSM schema to another party, and develop a formal framework that supports splitting and outsourcing of GSM schemas. One element of the framework focuses on restructuring the GSM schema to facilitate outsourcing while preserving the semantics of the original schema; the second focuses on locking protocols that define how the distributed parties should operate. Additionally, the framework allows parties to keep local parts of their GSM subschema private without affecting the outcomes of the global execution. The rules restructuring developed here enables a crisp separation of concerns, which allows reuse of existing GSM (and thus Case Management) engines for executing the subschemas. Both elements of the framework are formally proven correct.
ieee international conference on services computing | 2013
Roman Vaculín; Richard Hull; Maja Vukovic; Terry Heath; Nathaniel Mills; Yutian Sun
Collaborative Decision Processes focus on long-running, multi-faceted decisions (e.g., the planning and approval process for a municipal development project) made by teams of people. Such decisions can draw on the crowd in a variety of ways, by enabling large-scale on-line brainstorming, crowd-sourced queries, and text mining to determine public sentiments and derive new insights. Such decision processes consist of a combination of structured and highly unconstrained, human-drive activities. This paper describes a framework and a prototype system for support of collaborative decision processes based on declarative, data-centric dynamic artifacts and Advanced Case Management.
Confederated International Workshops on On the Move to Meaningful Internet Systems, OTM 2012: OTM Academy, Industry Case Studies Program, EI2N 2012, INBAST 2012, META4eS 2012, OnToContent 2012, ORM 2012, SeDeS 2012, SINCOM 2012, and SOMOCO 2012 | 2012
Yutian Sun; Wei Xu; Jianwen Su; Jian Yang
Business processes (BPs) can be designed using a variety of modeling languages and executed in different systems. In most BPM applications, the semantics of BPs needed for runtime management is often scattered across BP models, execution engines, and auxiliary stores of workflow systems. The inability to capture such semantics in BP models is the root cause for many BPM challenges. In this paper, an automated tool SeGA for wrapping BPs is developed. We demonstrate that SeGA provides a simple yet general framework for runtime querying and monitoring BP executions cross different BP management systems.