Service-Oriented Architectures for Smart Manufacturing Project

Objective:  Improve the performance of smart manufacturing systems by developing standards, reference models, and software tools that enable the application of service-based integration technologies to smart manufacturing systems.

Technical Idea:  Manufacturers are increasingly using the inherent power of cloud computing, resource virtualization, and other technological advances to increase agility and efficiency of their operations. These advances allow manufacturers and software providers to bring their own resources, data analysis, intelligence, and unique capabilities to be available for on-demand networking in the form of online services. Such online services are composed into a new generation of service-oriented manufacturing systems. These systems will be significantly more flexible and less expensive to use than the current generation of manufacturing systems made up of monolithic manufacturing applications. However, integrating heterogeneous services in the form of composite systems is not a trivial job. There is a need for service vendors, cloud vendors, manufacturers, and other stakeholders to work towards solving the problem of interoperability and composability of manufacturing services.

The primary technical barrier in achieving interoperability of manufacturing services is the lack of efficient, reliable methods to develop messaging standards. Messaging standards are an essential integration component in service-oriented manufacturing systems, enabling meaningful data exchange and information sharing via messages exchanged among services of the manufacturing system. Importance of messaging standards is becoming even greater as manufacturers participate in distributed eco-systems of manufacturing service providers, users, and platform providers. NIST discovered that the interoperability problem may be significantly reduced with new methods to develop messaging standards. The underlying technical idea is to address the absence of means to manage usage information of messaging standards. Usage information of a messaging specification is both (1) customized or profiled messaging standard and (2) context (intended usage) for the customized messaging standard. This would be derived through models of linkages between the standard and customized models of a messaging standard.

To address interoperability of services, the project will pursue the following research objectives:

• A new model-driven approach for developing manufacturing service messaging standards
• Industry standards necessary to represent and manage model-based messaging
• Measurement methods to determine verifiable, valid changes in messages

The primary technical barrier in achieving composability of manufacturing services is the lack of efficient, reliable methods to specify service compositions. While service interoperability is required for efficient service-oriented manufacturing systems, it is not sufficient. Interoperability assures correct interpretation of results between communicating services. Composability enables correct consequences of service interactions by using correct behavior models. Service composability is needed if ad-hoc compositions and re-configurations of manufacturing services is desired. Currently, manufactures use business processes as the foundation for integrating the right applications used to operate the manufacturing enterprise. Business process models act as blueprints of business processes and specify intended service behaviors. Our technical idea is that the business process models are appropriate basis to derive contextual information for specification of automated service composition. This will be done by developing business process meta-models and other reference models to allow derivation of valid business process models. This, in turn, allows for service behavior specification, required for the ad-hoc composition of services.

To address composability of services, NIST will pursue the following research objectives:

• Business Process (BP) model-driven manufacturing services composition approach
• Standards necessary to represent and manage BP model-driven services composition; and measurements to determine verifiable, valid changes in BP models for services composition.

Research Plan:  To address interoperability of services objectives, the project will conduct the following research activities:

• Model-driven development approach for manufacturing service messages
  • Developing a relational model for syntax-independent messaging standard based on the CCS (ISO 15000-5) standard;
  • Designing an innovative collaborative messaging standard development processes.
  • Building messaging standard development and deployment tool allowing for specification of profiled messaging standards, generation of various syntax-specific representation of messaging standards, and the collaborative messaging standard development;
• Standards necessary to represent and manage model-based messages by
  • developing a usage-based message life-cycle management method;
  • developing a standard for representing the messaging standard usage information.
• Measurements to determine verifiable, valid changes in messages by
  • developing verification tests with respect to the devised relational model; and
  • developing algorithm for detecting and suggesting changes.

These activities will culminate in the development and publication of the Messaging Standard Semantic Refinement Tool (MSSRT) described below.

To address composability of services objectives, the project will conduct the following research activities:

• Business process (BP) model-driven manufacturing service composition approach
  • developing business process meta-model based on ebRIM (ISO/TS 15000-3) standard;
  • advancing manufacturing business process modeling and meta-modeling approaches;
  • designing business process model-based context specification approach.
• Standards necessary to represent and manage BP model-driven service composition
  • developing standard development methods to specify context for usage-based (1) message profiles, and (2) business processes.
• Measurements to determine verifiable, valid changes in BP models for service composition
  • computational representation of business process context information;
  • semantic reasoning, use of rules and patterns to detect problematic changes.

These activities will culminate in the development and publication of the Business Process Cataloging and Classification/Contextualization System (BPCCS) described below.

We will pursue our research plan by working with industry on building three new software tools to develop and test our new methods for interoperable and composable services: (1) Messaging Standard Semantic Refinement Tool (MSSRT) which addresses interoperability by improving and automating the standards development process; (2) Business Process Cataloging and Classification/Contextualization System (BPCCS) will address composability issues by linking business process models to a common meta-model; and (3) Real Time Service Oriented System (RTSOS) tool that uses our results in composability and interoperability at the enterprise level to extend these capabilities to lower-level manufacturing systems that have time dependencies.

MSSRT is being developed initially for the OAGIS data exchange standard to aid the service providers and users in generating and cataloging the messaging standard usage information using a new, CCS-compliant OAGIS meta-model. This tool will be used as a proof-of-concept (POC) web-based application software environment for the life-cycle management of messaging standards, including the usage information. In parallel, the tools will be utilized to experiment with new methods to create, maintain, and use messaging standards, leading to industry validation. The POC software tool will validate planned collaborative, multi-tenant methods and meta-models for life-cycle management of messaging standards. It will provide core functionalities that can be extended and commercialized by industry.

BPCCS is being developed to allow automated generation of both standard and context-specific OAGIS business processes. A core focus of BPCCS will be to provide to open community of stakeholders shared definitions of required concepts and terms for business processes that span across an enterprise and its multitier supplier network. This tool will enable context information life-cycle management in support of the management of usage information for the MSSRT tool as well for business process model discovery and reuse. We will design the tool to consist of two main components: (1) a catalog, for storing business process models (including reference models); and (2) classification scheme, for classifying process models stored in the catalog. The classification scheme is the core enablement for capturing the context for an intended usage of the models (e.g., specifying functional actors and their intentions that trigger and receive required integration messages). Both components will be developed to manage the lifecycle of their content.

RTSOS is being developed to allow flexible and dynamic interoperation of autonomous software and manufacturing systems at shop floor level. We will develop modeling techniques and tools for enabling the (semi-) automatic composition and analysis of these services. RTSOS development will be focused on reference model of real time manufacturing services, which will lead to a RTSOS tool development to allow generation of messaging standards for real-time service-oriented modeling and analysis. The tool will enable process control information modeling for various manufacturing industries. RTSOS will provide a meta-model that specifies requirements for flexible, service oriented manufacturing systems. Based on the common meta-model and open data access management, process independent intelligent applications can be developed and configured quickly for various manufacturing systems, ranged from continuous production to discrete factory assembly.

In addition to working with industry consortia, we will continue to organize annual R&D workshops on Service-Oriented Manufacturing, focused on enabling interoperable and composable manufacturing services. We plan to extend the reach of the annual workshop series and use it as part of our collaboration and our research-product transition strategy.