Silviu Răileanu

Switching mode control strategy in manufacturing execution systems

The paper describes a semi-heterarchical control solution for mixed planning and scheduling, routing and job execution in flexible manufacturing systems based on the paradigms of holonic manufacturing and product-driven automation. The main feature of the control solution is the bidirectional switching of the operating mode (scheduling, routing) between centralised and decentralised in the presence of perturbations to ensure as long as possible both global optimisation and agility to changes in batch orders, while featuring robustness to disturbances in the production environment. At the theoretical level, the control solution is described in terms of generic structural and dynamic models. The implementation is done using a multi-agent Java Agent Development Framework framework.

A Service Oriented Architecture for Total Manufacturing Enterprise Integration

The paper describes a, framework and implementing issues addressing service orientation in the management and control of enterprises. A hybrid, semi-heterarchical control model based on the paradigm of holonic manufacturing is proposed; it switches its operating mode from hierarchical to heterarchical in the presence of perturbations to ensure both global optimization and agility to changes in batch orders, while featuring robustness to disturbances in the production environment. In order to ensure these conflicting functionalities, a service oriented architecture is proposed and implemented, whose structure includes a distributed fault-tolerant Resource Service Access Model. Besides the design of a generic structural and dynamic model, a real implementation solution is proposed using a multi-agent framework.

Centralized HMES with Environment Adaptation for Production of Radiopharmaceuticals

The paper presents an intelligent hybrid control solution for the production of radiopharmaceuticals that irradiates radioactive isotopes of neutron-defficient radionuclides type by in cyclotrons. To achieve requirements such as: highest number of orders accepted daily, shortest production time and safe operating conditions, a hybrid control system based on a dual architecture: centralized HMES with ILOG planning and decentralized parameter monitoring and control via SCADA is developed. Experimental results are reported.

Robotic 3D Surface Laser Scanning for Feature-Based Quality Control in Holonic Manufacturing

The paper describes an on line geometry control technique which uses a laser sensor hand-held by a 6 d.o.f.-vertical articulated robot and displaced along predefined scanning paths. The 3D surface of manufactured part is digitally reconstructed after the robot drives the laser sensor along a scanning motion pattern corresponding to the part type (prismatic, revolute) and geometric features to be checked (direct measurements: lengths, angles, signatures, depth map evaluation: local and global indicators) against reference part models. The robotized laser scanning system is integrated with a 2D vision platform allowing thus to perform an extended set of geometry inspection tasks on manufactured parts (processed on CNC machine tools, assembled by robots). The described quality control technology is applied to holonic manufacturing by including in the order holon the inspection sequence retrieved from the product holon and the laser scanner robot enabled by the resource holon. Implementing solutions and experimental results are reported.

Manufacturing Systems at Scale with Big Data Streaming and Online Machine Learning

Real time analysis of data collected from the shop floor opens the path towards efficient scheduling of batch execution for large scale distributed manufacturing systems. Prediction of the shop floor activities has a great potential to reduce manufacturing costs, by providing the information required for operational decisions like preventive maintenance, automatic remediation or scheduling optimization. Research has been focusing on how machine learning algorithms can be used to better understand and extract insights from historical data collected from manufacturing systems. However, in the current manufacturing environments, driven by mass customization and short time to market, these approaches fail to be agile enough to be useful. In this paper we propose a real-time machine learning approach for large scale manufacturing systems that can predict various scenarios before service degradation occurs, thus allowing for corrective actions. At the same time, outliner detection algorithms can be used to evaluate the system’s health at a holistic level. Scalability requirements are achieved by modelling the architecture around data streams processed in real time by map-reduce operations. The concepts presented in this paper build on recent developments on flexible, distributed and cloud based manufacturing, where these real time actions can be efficiently implemented.

Service Orientation of Environment Control Processes

The paper introduces a framework for designing flexible Environment Control Services (EServices) based on generic sensing, modelling and control process specifications allowing the customization of distributed Facility Environment Control Systems (HFES) in holonic approach. The representation, assessment, configuring and implementation of these EServices by the interacting holons using data structures models in different perspectives their type, features, functionalities, interdependencies and usage: (i) the EService Type perspective used to identify a requested type of service included in an existing service-ontology; (ii) the EService Specification perspective allowing the client (the environment conditioning application) to define the facility’s environment needs as service properties; (iii) the EService Profile perspective in which the resources publish their capabilities matching the requested EService Type, exposing them to (iv) the EService Configuring and Implementation perspective. In the resulting Service-oriented HFES (SoHFES), four types of processes: environment monitoring, control, conditioning and production influencing, represented in the service perspective either as simple or composite services according to timing progression, granularity and concurrency of their operations can be standardized into EServices which allows for flexibility, agility, robustness and reusability of control solutions. A generic design framework and experimental results are finally presented.

Dynamic Service Capacity and Demand Matching in a Holonic Public Transport System

The paper describes the development of a fleet management system for public transport services, operating in semi heterarchical mode: (1) hierarchical on long term, with optimal activity planning and resource scheduling; (2) heterarchical– agile and robust at disturbances, with real-time rescheduling of unavailable drivers, not operational busses and blocked routes. The holonic paradigm is used; the holonic transport control architecture, the types of holons and the basic cooperation rules between them are presented. The paper describes the Service Scheduler, as centralized, hierarchical information system used for optimized, long-term capacity managing. The decentralized, heterarchical Delegate MAS (Multi-agent System) performs real-time change management for capacity and demand matching, based on intelligent devices embedded on resources. An implementing MAS framework based on JADE is proposed.

Service Oriented Mechanisms for Smart Resource Allocation in Private Manufacturing Clouds

Cloud manufacturing (CMfg) represents an evolution of networked and service-oriented manufacturing models, focusing on the new opportunities in networked manufacturing enabled by the emergence of cloud computing platforms. The paper reports results obtained in developing service-oriented mechanisms that optimize resource allocation in a private cloud implemented with IBM CloudBurst 2.1 for manufacturing enterprises. A predictive mechanism is described that augments the generic threshold-based implementation by recognizing repetitive patterns in application usage. Then, a queued resource allocation bus architecture extended with computation of alternative schedules is proposed; this is a service oriented mechanism assuring adaptability of the private CMfg system to workload fluctuations; the implementation is on IBM CloudBurst 2.1 and uses a genetic algorithm for smart resource allocation. A multilayer QoS monitoring architecture, designed for the cloud, is reported.

Holonic Facility Environment Monitoring and Control for Radiopharmaceutical Agent-Based Production

The paper presents design and implementing considerations for an agent-based environment monitoring and control system dedicated to a radiopharmaceuticals production line. This system is distributed on two layers: a centralized one, responsible for updating the status of the facility’s parameters and the global control strategy for cascade room parameters, and a distributed one that applies the global strategy to local environment control units—Heat Ventilation and Air Conditioning units (HVAC). The paper describes the environment parameters and defines the HVAC process models. The facility environment control system is developed in holonic approach, with three basic holons and one expertize holon. An implementing solution and integration of the environment control system with the radiopharmaceutical production management system is described. Experimental results and conclusions are finally presented.

Product-Driven Manufacturing Control with Embedded Decisional Entities

Abstract The paper describes a 2-layer agile holonic control architecture designed for optimizing production on both long and short term. The proposed planning, scheduling and control model is generic, the structuring of decisional entities and the distributed decision making do not rely on proprietary technologies. An implementing framework for product-driven manufacturing control based on intelligent embedded devices and the JADE multi-agent platform is given in the paper. The proposed 2-layer open control model with semi heterarchic mode switching between hierarchical and heterarchical behaviour demonstrates how Multi-Agent Systems and Holonic Manufacturing Execution Systems combined usage provides sustainable manufacturing through service orientation. The high availability service access model developed brings closer the physical and decisional parts offering global performance and agility.