[1]李修贤,王桂玲,石永鹏.面向主动式 BPM 的 IoT 服务动态绑定方法[J].计算机技术与发展,2023,33(09):64-71.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 010]
　LI Xiu-xian,WANG Gui-ling,SHI Yong-peng.A Dynamic Binding Method for IoT Services towards Proactive BPM[J].,2023,33(09):64-71.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 010]

点击复制

面向主动式 BPM 的 IoT 服务动态绑定方法()

分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

33

期数:

2023年09期

页码:

64-71

栏目:

移动与物联网络

出版日期:

2023-09-10

文章信息/Info

Title:

A Dynamic Binding Method for IoT Services towards Proactive BPM

文章编号:

1673-629X(2023)09-0064-08

作者:

李修贤¹; 2 ; 王桂玲¹; 2 ; 石永鹏¹; 2

1. 北方工业大学 信息学院,北京 100144;

2. 北方工业大学 大规模流数据集成与分析技术北京市重点实验室,北京 100144

Author(s):

LI Xiu-xian1; 2 ; WANG Gui-ling1; 2 ; SHI Yong-peng1; 2

1. School of Information,North China University of Technology,Beijing 100144,China;

2. Beijing Key Laboratory on Integration and Analysis of Large-scale Stream Data,North China University of Technology,Beijing 100144,China

关键词:

业务流程管理; 情境感知; 物联网服务; Seq2Seq 模型; 注意力机制

Keywords:

business process management; situation-awareness; IoT services; Seq2Seq model; attention mechanism

分类号:

TP311

DOI:

10. 3969 / j. issn. 1673-629X. 2023. 09. 010

摘要:

在现有的物联网(IoT) 服务及业务流程管理(BPM) 系统研究的基础上,提出一种面向主动式 BPM 的 IoT 服务动态绑定方法。 首先,该方法构建了从时序序列中学习情境信息及 IoT 服务绑定规律的模型,并预测可绑定的 IoT 服务列表。该模型基于双向门控循环单元( Bi-GRU) 搭建编码器解码器结构网络模型,引入注意力机制,通过优化神经网络结构提取时序数据依赖关系,从而通过该预测模型提升 BPM 系统的主动性。 其次,将用户任务与边界非中断事件相结合实现 IoT服务动态绑定活动,该活动通过集成上述预测模型和 IoT 服务动态切换及多 IoT 服务并行运行架构实现 BPM 系统在运行时动态绑定 IoT 服务。 实现面向主动式 BPM 的基于情境感知的 IoT 服务动态绑定系统,基于液化天然气( LNG) 危化品海上运输安全监管场景进行案例分析,对系统中预测模型的性能进行了定量的实验评价。 经过对比,提出的预测模型在多项指标中均优于其他模型。 案例分析和实验结果验证了方法的有效性。

Abstract:

A dynamic binding approach of IoT services for proactive BPM is provided based on the existing research on Internet of Things( IoT) services and business?

process management ( BPM) systems. Firstly, such method constructs a model for learning situationalinformation and IoT service binding laws from temporal sequences and predicts the list of bindable IoT services,which builds an encoder-decoder structure network model based on bi - directional gated recurrent units

?( Bi - GRU) , introduces an attention mechanism, andextracts temporal data dependencies by optimizing the neural network structure,so as to enhance the BPM system’ s proactivity throughthis prediction model. Secondly,combining user tasks with border non-interruptible events to achieve dynamic binding activities for IoTservices,this activity enables the BPM system to dynamically bind IoT services at runtime by integrating the aforementioned predictivemodel with IoT service dynamic switching and multiple IoT services running in parallel architecture. Situation - aware IoT servicedynamic binding system for proactive BPM is implemented.?

A quantitative experimental evaluation of the performance of the predictionmodel in the system is carried out based on a case study of a safety supervision scenario?

for the marine transportation of liquefied naturalgas ( LNG) hazardous chemicals. After comparison,the proposed prediction model outperforms other models in several indicators. Thecase study and experimental results verify the effectiveness of the proposed method.

常用功能