Collaborative Control Theory Research Articles

Preemptive Demand and Capacity Sharing Learning Protocols Using Long Short-Term Memory (LSTM) Neural Network Autoencoders

This research is conducted as a contribution to the design of future resilient global supply, subject to disruptions. In a collaborative network of enterprises (CNEs), companies laterally collaborate through information and resource sharing. A common practice of collaboration is through demand and capacity sharing (DCS), which aims to enable timely delivery to customers, despite uncertainties in the market and disruptions in available supply capacity. One key challenge of DCS implementation is to ensure that capacity-sharing proposals from one enterprise to another are generated efficiently. A slight delay in DCS protocols could lead to collaboration failures and backorder fees, prevalent if companies initiate the protocols when actual demand has been presented. This study aims to develop a new preemptive DCS, following CCT, Collaborative Control Theory protocol for DCS, in which machine learning is utilized to learn from historical market data and invite capacity-sharing proposals before actual demand occurs. Long Short-Term Memory (LSTM) Neural Network Autoencoders are used to detect anomalies in real-time. The detected anomalies become enablers of the capacity sharing proposal, which could initiate DCS preemptively. A case study of automobile manufacturers in Indonesia is used to validate the model. On average, for the given case study, preemptive protocols based on the learning protocols are capable of increasing the fulfillment rate to 91.03% in the demand-capacity sharing CNEs; better compared to the non-collaborative and non-preemptive demand-capacity sharing protocol (DCSP) baseline scenarios. Further research is being conducted for refinements.

From Student’s Experience: Does E-learning Course Structure Influenced by learner’s Prior Experience, Background Knowledge, Autonomy, and Dialogue

Background: E-learning is increasingly becoming a preference in higher education institutions worldwide; this is intended to assist educational institutions in achieving objectives to meet the proportion of individuals with their educational opportunities. Nevertheless, instructors and students frequently have concerns with their capacity to succeed in E-learning environments. Objectives: This study aimed to presents common eLearning challenges in regard to e-learning courses structure and its relations to various factors, for instance; students’ autonomy, prior knowledge and experience, students- students dialogue, and students- instructor dialogue, and proposes solutions to these challenges based on the transactional distance theory. Moreover, this study presents evidence from Malaysian higher institutions based on theoretical models for e-learning course structure and its relations to the factors mentioned above. Methods: Data have been collected from 680 university learners all over Malaysia. Data were then examined using exploratory factor analysis, confirmatory factor analysis, and structural equation modelling employing Smart PLS 3.0 software. Results and conclusion: Research findings indicated that e-learning course structure was affected by all dimensions of overall path analysis findings: student autonomy, students background, student-instructor dialogue, and student-student dialogue. However, the e-learning course structure showed insignificant with students’ prior experience. Implications: Implications for universities are discussed accordingly. Such findings provide vital support to the integrative association among collaborative control (CC) and transactional distance theory (TDT) regarding e-learning environments experience, which might support universities administrators in the higher education industry to implement, plan and evaluate online learning platforms applications in their institutions.

Dynamic-distributed decisions and sharing protocol for fair resource sharing in collaborative network

This research addresses distributed decisions for dynamic partnership reformation to enhance fairness and efficiency among collaborative networked enterprises (CNEs). The aim of this research is to investigate a new sharing protocol (SP) to control CNEs. Most CNE studies are designed under a centralized SP. However, centralized SPs appear to be unfeasible or inadequate due to enterprises’ operating environments and local objectives; therefore, each CNE requires a distributed decision-making mechanism. Furthermore, market globalization and competition necessitate rapid responses to market changes and give rise to two major concerns about CNEs: (1) how can collaborations be rendered re-configurable to dynamically capture and adapt to market patterns; and (2) how can the resources of all CNEs be utilized fairly among enterprises, ensuring mutual benefits? Therefore, Dynamic-Distributed Sharing Protocol (D2SP), which is inspired by the principles of Collaborative Control Theory and Contract Net Protocol, is proposed to address these concerns. The performance of D2SP is evaluated by comparing with (1) static-centralized SP, (2) static-distributed SP, and (3) dynamic-centralized SP. The experimental results show that D2SP performs at least equally or better than centralized and static SPs in terms of total profit, fairness, scalability, and communication overhead.

A collaborative control protocol for agricultural robot routing with online adaptation

Stress in crops is one of the major concerns in precision agriculture because it indicates the emergence of disease and damage in plants. Detecting the stress condition of a plant early is critical. A system that can monitor the condition of plants is a desirable solution. In this work, Collaborative Control Theory is utilized to construct a new system, ARS (agricultural robotic system) which synchronizes humans, a mobile robot, and a variable set of sensors to effectively perform the monitoring and detection tasks. A key protocol for that system, which combines routing algorithm, adaptive search algorithm, and collaboration control framework has been developed and validated, and is presented in this article. By using greenhouse as a case study structure, the protocol routes a robot to visit the sampled locations by using a genetic algorithm. In addition, the search algorithm can be guided by the predictive characteristics of the crops’ stress, which can spread to other plants according to sunlight, airflow direction, and other known conditions. Based on simulation experiments, the results indicate with statistical significance that (1) the routing algorithm increases the number of successful detections of existing stressed plants by 45.77% compared to monitoring without this routing algorithm. (2) The adaptive search algorithm improves the number of successful detections of stressed plants by 71.88% compared to a system without the adaptive search algorithm. (3) The new protocol developed in this research yields the highest overall robotic efficiency, compared with a system without collaborative control framework.

A two-phase coordinated logistics planning approach to platelets provision in humanitarian relief operations

For provision of platelets in humanitarian relief operations, due to the excessive number of non-coordinated organizations, relief response times and corresponding costs of platelets sharply increase. This article proposes a two-phase mechanism to coordinate two heterogeneous relief organizations (i.e., relief and rescue organization of the Red Crescent Society and a blood transfusion organization) in a decentralized network in such a way that its own interests and objectives are also satisfied. The blood transfusion organization tries to minimize the wastage level of platelets considering related total costs. At the same time, the relief and rescue organization decides on the selection of shelters where the platelets can be administered to injured people, while minimizing total relief time. Thus, first a bi-level mixed integer linear model under the demand and supply uncertainties is developed (phase 1), and then a capacity-sharing coordination mechanism based on the collaborative control theory is proposed (phase 2). To solve large-scale instances, a fuzzy Kth-Best algorithm is developed to solve the first phase and then phase 2 is solved by the proposed coordination mechanism. We compare our model to centralized relief logistics model using a data set for a possible earthquake in Tehran, Iran. Results show that our model reduces shortage and wastage compared with the centralized model

Collaboration Requirement Planning Protocol for HUB-CI in Factories of the Future

Rapid advances in production systems’ models and technology continually challenge manufacturers preparing for the factories of the future. To address the complexity issues typically coupled with the improvements, we have developed a brain-inspired model for production systems, HUB-CI. It is a virtual Hub for Collaborative Intelligence, receiving human instructions from a human-computer interface; and in turn, commanding robots via ROS. The purpose of HUB-CI is to manage diverse local information and real-time signals obtained from system agents (robots, humans, and warehouse components, e.g., carts, shelves, racks) and globally update real-time assignments and schedules for those agents. With Collaborative Control Theory (CCT) we first develop the protocol for collaborative requirement planning for a HUB-CI, (CRP-H), through which we can synchronize the agents to work smoothly and execute rapidly changing tasks. This protocol is designed to answer: Which robot(s) should perform each human-assigned task, and when should this task be performed? The primary two phases of CRP-H, CRP-I (task assignment optimization) and CRP-II (agents schedule harmonization) are developed and validated for two test scenarios: a two-robot collaboration system with five tasks; and a two-robot-and-helper-robot collaboration system with 25 tasks. Simulation results indicate that under CRP-H, both operational cost and makespan of the production work are significantly reduced in the two scenarios. We summarize with the implications and future plans for integrating HUB-CI and CRP-H in a cyber-augmented physical simulation model.

Collaborative Control Protocol for Agricultural Cyber-physical System

In this article, a Collaborative Control Protocol for Robotic and Cyber-Physical System (CCP-CPS) is presented. CCP-CPS, which enables an application of robotics in a CPS system for smart and precision agriculture (PA), aims to monitor and identify stresses in greenhouse crops by a hyperspectral analysis method. Roles of CCP-CPS are to assign tasks to agents, identify and resolve conflicts and errors in the system, and enable a more effective collaboration and interaction among agents in CPS, compared to traditional non-collaborative and non-CPS approach. Collaborative Control Theory (CCT) is utilized for two system levels; protocol and agent level (CCP level), and CPS and environment level (CPS level). We use two case studies to test and validate our method with alternative approaches. The results from computer simulation show that 1) CCP-CPS can identify the highest number of greenhouse locations containing stresses, 2) CCP-CPS can utilize available resources (time) effectively, and 3) CCP-CPS can respond to an emergency stress situation faster and have more tolerance capability with system conflicts and errors, by having human integration and cyber-augmented greenhouse system, than other monitoring systems.

Agricultural cyber physical system collaboration for greenhouse stress management

This article presents a CPS (Cyber Physical System) oriented framework and workflow for agricultural greenhouse stresses management, called MDR (Monitoring, detecting and responding)-CPS. MDR-CPS has been designed to focus on monitoring, detecting and responding to various types of stress. CCT (Collaborative Control Theory) is applied in MDR-CPS to deploy CRP (Collaborative Requirements Planning), address CEs (Conflicts and Errors) and enable a collaborative architecture for better CPS interactions. Analytic studies and simulation experiments are conducted to compare our scheme with an alternative scheme for agricultural greenhouse stress management. The results show that MDR-CPS performance is better than the compared scheme, in terms of detection cooperated with human, CEs tolerance, and emergency response. Focus in this design and development is on the collaboration among sensors, robot, and human to improve the MDR-CPS’ performance.

The constrained-collaboration algorithm for intelligent resource distribution in supply networks

Manufacturing and supply strategies have evolved from the notion of mass production focusing on economies of scale, to flexible production systems seeking economies of scope, and recently, to the concept of enterprise and supply networks aiming for economies of collaboration. The emergence of supply networks poses new challenges derived from a growing complexity in coordinating the flow of resources, materials, and information within and among an increasing number of network participants. In such situations, an intelligent resource distribution under various constraints is one of the most critical problems. In this paper, we have developed a novel Constrained-Collaboration Algorithm (CCA) for physically cooperative resource distribution planning. Based on Collaborative Control Theory (CCT) with theoretical formulations and network flow approaches, the CCA addresses an efficient and effective resource distribution by a suitable form of physical cooperation. We have applied it to an actual industry case, in combination with Direct/Indirect Delivery Protocol (DIDP), introduced in previous research. As a result, the integrated model achieves 55% increase in resource utilization and 20% reduction in distribution cost, while accommodating external changes. Besides, the new formalism of the physical dimension of collaboration requirement planning introduced in this research with new CCA approach can be generalized for other improvements in supply-and-demand management decision support.

The collaborative factory of the future

This work is a contribution to the ever-evolving concept of factory of the future (FoF); thinking about its challenges, and how we can best accomplish its promise. Globalised economies, competitive, volatile and complex business environments have gradually transformed modern manufacturing processes into highly distributed but interconnected networks of entities, which divide their responsibilities and core competencies to sustain and evolve. To achieve these ends, collaboration among networks of intelligent and autonomous entities must be effectively engineered and augmented. In view of that need, a formalism of the collaborative control theory (CCT) is developed in this article, based on the state-of-the-art advances in its design principles, supported by collaborative e-Work and their roles in engineering augmentation of collaboration. Several specific examples of e-Manufacturing systems (e-Factories) are used to elaborate further the mechanics of the developed formalism. The goal is to provide a framework of the collaborative factory of the future (CFoF). This framework is applied to enable the abstraction of disparate e-Factories as collaborative networks of entities, which in turn can be engineered and augmented through the CCT operators in order to address the FoF-2020 challenges of scalability and sustainability.

A resilience by teaming framework for collaborative supply networks

Supply network resilience is an emerging concept related to the ability of a network to tolerate disruptions; current understanding of its meaning and dimensions, its role in the design and operation of supply networks, and its relation to sustainability is at its early stages. Existing approaches are based on the trade-off between increased resources and higher fault-tolerance. The Fault Tolerance by Teaming (FTT) principle of Collaborative Control Theory has been applied in sensor networks effectively and appears as a promising original approach not based on the aforementioned trade-off and capable of producing networks with higher resilience.Inspired by the FTT principle, a Resilience by Teaming Framework (RBT) for supply networks is developed to address the design and operation of resilient supply networks. RBT is tested and validated through the application of its protocols to case studies in production and distribution networks. Evidence from case studies’ results suggests that through FTT-based protocols and RBT it is possible to achieve higher fault tolerance with fewer resources than under traditional approaches.

Adaptive direct/indirect delivery decision protocol by collaborative negotiation among manufacturers, distributors, and retailers

Recently, as the dynamic nature of supply networks increases, the importance of adaptive design, planning, and management becomes more emphasized. Especially, as the complexity of interacted distributed operations increases under changeable conditions, it becomes a key issue to design the adaptive collaboration strategy considering the multiple network participants׳ economic goals and interdependent factors that affect/restrict their viability. Hence, this paper addresses one of related supply network problems—an adaptive direct/indirect delivery by using appropriate collaboration to deal with changeable conditions. We have developed a Direct/Indirect Delivery Protocol (DIDP), which aids to model physically and economically viable collaboration strategies and provides effective delivery schedule under changeable conditions. This decision protocol is designed based on Collaborative Control Theory (CCT), and includes Delivery Plan Generation sub-protocol (DPGsp) for delivery pattern estimation and optimization process. To evaluate its performance, a numerical example is conducted; it results in accommodating to external changes well with 55% increase in the resources utilization ratio and 20% reduction in the total physical distribution cost, compared to original delivery pattern.

Balanceable assembly lines with dynamic tool sharing and best matching decisions—a collaborative assembly framework

A Collaborative Assembly Framework (CAF), inspired by the design principles of the Collaborative Control Theory, is developed in this article to enhance the extent of balancing of assembly lines. The notion of the CAF lies in the dynamic utilization of idle resources to eliminate bottlenecks. The CAF is composed of two modules: the first one, the Tool Sharing Protocol (TShP), makes dynamic tool-sharing decisions among fully loaded (i.e., bottleneck) and partially loaded Work Stations (WSs), and the second one, the Best Matching Protocol (BMP), dynamically matches tasks and WSs (BMP-1) and partially and fully loaded WSs for tool sharing (BMP-2). A Multi-Objective Mixed-Integer Programming model is developed for mathematical representation and a Fuzzy Goal Programming approach is applied for optimization purposes. The objectives are to minimize (i) the number of WSs, (ii) cycle time, and (iii) the total collaboration cost. The developed CAF is proven to guarantee the relative extent of balancing of assembly lines, depending on pairwise tool compatibility and tool-sharing performance. Numerical experiments on a set of small-sized case studies repeated and expanded from previous research show the superiority of the CAF over the existing non-collaborative approaches in terms of line efficiency, utilization, and balancing.

Collaborative Intelligence - Definition and Measured Impacts on Internetworked e-Work

Abstract Internetworked e-Work is enabling new channels in cyber space for collaboration among physical and cyber agents, e.g., humans, robots, software agents. Research on Collaborative Control Theory (CCT) indicates that building and augmenting the Collaborative Intelligence (CI) of participants in cyber-physical networks can provide better support for achieving their individual and common goals. In spite of its rising significance and popularity, however, no clear and precise definition and universal quantitative measure has been proposed for the CI. In this article, we first formalize the CI by suggesting a formal definition, based on the definitions of its elements - collaboration and intelligence. We then propose a quantitative measure for the CI, adapted from the universal intelligence measure. For illustration, we analyze three recent collaborative e-Work studies at three different scales: (1) Telerobotenabled computer supported collaborative design; (2) Collaborative product line control in supply networks; (3) Demand and capacity sharing in multi-enterprise collaboration. From these case studies, common advantages such as work efficiency, network robustness and stability, service level, resource utilization, and collaboration cost are observed, analyzed, and translated into formal CI measures. Results indicate significant impacts of CI on the efficiency, effectiveness, and quality of collaborative activities in emerging e-Work networks.

Asynchronous cooperation requirement planning with reconfigurable end-effectors

Automation processes in an unstructured environment are challenged by changing peripheral requirements. Reducing the cost of peripherals is a major concern for system designers. Addition of extra flexibility to the existing equipment to handle a larger range of tasks is a desirable solution, which can be offered by reconfigurable end-effectors (REEs). A REE system has an adjustable structure to facilitate the adaptation of the end-effectors to various objects, so it is an intermediate solution between flexible and dedicated end-effectors. To balance the trade-off between the production quality and the reconfigurability becomes the main difficulty in the design and control of REE. In this work, the models of configuration network and order of reconfigurability are introduced to provide theoretical analysis of the reconfigurability in REE. Based on the analytical models, the asynchronous cooperation requirement planning (ACRP) framework is established to facilitate the effective design and control of REE. ACRP provides a dynamic solution extended from the planning facet of collaborative control theory (CCT) for designing (offline) and controlling (online) multi-agent collaborations. ACRP determines how flexible an automation system should be considering the cost of reconfigurations and the quality of production. The framework is illustrated with a case study of vegetable harvesting by multi-arm automated systems. In harvesting processes, the grasp quality is one of the most important factors for production quality. Simulation experiments in the current article show that both new and previous CCT methods increase average grasp quality of harvesting by 14% compared to harvesting without CCT. Furthermore, the new framework, ACRP, outperforms the previous CCT method, by a 7% improvement of the total harvest yield. Design of reconfigurable end-effectors system for flexibility and cost-savingsAnalyze reconfigurability based on configuration network and its structureBalance the trade-off between the production quality and system reconfigurabilityControl reconfigurable system by collaborative multi-agent asynchronous planningCase study on grasping quality and yields in robotic harvesting

Real-time administration of tool sharing and best matching to enhance assembly lines balanceability and flexibility

Collaboration has been found in previous studies on the design of assembly lines to be a useful mechanism. In this study, the focus is on a collaborative assembly (CA) framework, inspired by the design principles of CCT, the Collaborative Control Theory, to improve balanceability and flexibility of assembly lines through tool sharing (TS) among idle and bottleneck workstations. TS is widely practiced in advanced assembly facilities to reduce cost and improve consistency and standardization in assembly and in assembly-and-test utilities, relying often on real time control. The framework developed here addresses the systems design aspect of Mechatronics, covering the planning, execution, and control mechanisms. Planning includes assembly line balancing (ALB) and initial TS decisions, made continually by solving a bi-objective mixed-integer program (BOMIP). A collaborative multi-agent system (CMAS) enhanced with a TS-best matching (BM) protocol is developed to execute the plan, control the process, and modify the TS decisions, considering dynamic changes in the system’s operations. Experiments show that the new CA framework significantly outperforms classic approaches (i.e., ALB without TS-BM) in terms of cycle time, utilization of tools, and balanceability. In addition, the control mechanism is proven to augment the line’s flexibility against the inherent uncertainties of assembly processes, compared to the previously developed static CA frameworks.

Resilience by teaming in supply network formation and re-configuration

In an increasingly interconnected world, physical, service, and digital supply networks are becoming progressively more complex, dynamic, and interdependent. Moreover, sustainability concerns push for higher efficiency in the use of resources, reducing protective redundancies and making supply networks more susceptible to disruptions. Supply network resilience is an emerging concept related to the inherent ability of a network to tolerate and overcome disruptions, a central capacity to achieve long term-sustainable operations. However, current understanding of the impact of network formation and re-configuration mechanisms on resilience is at its early stages.In this article, resilience by teaming (RBT) association decisions, inspired in the fault-tolerance by teaming principle from Collaborative Control Theory, are characterized and applied in network formation and re-configuration mechanisms. The impact of RBT on emerging supply network topology and resilience under targeted and random disruptions is measured and compared to scenarios were only traditional association rules are applied. Results show that the introduction of supply network agents using RBT association rules significantly (1) increases quality of service (QoS) with no significant cost increase in normal operation, (2) increases post-disruption QoS by 15–30% under random disruptions and 10% under targeted disruptions without cost increase vs. normal operation - when at least 20% of the agents use RBT association rules.

Collaborative production line control: Minimisation of throughput variability and WIP

Production research challenges over the last four decades are reviewed retrospectively. Industries can benefit from collaborative production advancements. Specifically, industries with automated, discrete production lines struggle with intrinsic process variability affected by rules that coordinate interactions among production line machines. Benefits can be gained by responsive collaboration among these machines. A given set of rules can potentially yield either an amplifying or smoothing effect on the process’s inherent variability, subject to current system conditions. A key challenge is the lack of models to dynamically select/re-select the appropriate set of rules to optimise performance.This problem involves uncertain and dynamic constraints, raising the need for collaboration and conflict resolution mechanisms. Through collaboration and conflict resolution the production flow disruptions can be minimised; high process efficiency to meet customer demand with minimum inventory levels can be achieved. A collaborative control theory approach based on three of its principles is developed to accomplish this objective. Performance under this approach, called collaborative production line control (CPLC) is compared to common industry standard control methods. A case study shows that CPLC outperforms traditional models, achieving a statistically significant 43% reduction in throughput variability, 24% reduction in WIP, and a 0.45% increase in service level.

Dynamic Tool Sharing with Best Matching Protocols for Efficient Assembly Line Balancing

In a typical assembly line, workstations can be classified into two sets: A) Fully loaded ( i.e. , bottleneck) and B) Partially loaded, with respect to the total processing times of their tasks and the cycle time. Collaborative Control Theory (CCT) is applied in this paper for efficient Assembly Line Balancing (ALB). Collaboration is carried out through: 1) Dynamic Tool Sharing (TS) between workstations in Sets A and B, and 2) Dynamic Best Matching (BM) between tasks and workstations. The purpose: Minimize the number of workstations and the cycle time, thus maximizing the line utilization/efficiency and the production throughput, and increase capability in dealing with demand variability. Collaboration between workstations is enabled, under TS protocol, in order for the workstations in Set A to make use of idle tools of workstations in Set B, thus minimizing the total processing time of the bottleneck workstation(s) ( i.e. , cycle time). A dynamic BM protocol is designed for optimal matching of tasks to workstations with respect to their processing times, precedence relations, and the cycle time. A bi-objective Mixed-Integer Programming (MIP) formulation is developed for modeling and analyzing the ALB-TS-BM decisions, minimizing the number of workstations ( i.e., Type -I ALB), and the average cycle time ( i.e. , Type -II ALB). A Fuzzy Goal Programming (FGP) method with imprecise goal hierarchy is applied for solving the bi-objective MIP model. The ALB-TS-BM model is original and first ALB solution method to consider CCT by TS, generalizable to and implementable in various instances of ALB problems through proper application of CCT principles. An illustrative example is used to exemplify the concept and its advantages over conventional ALB methods.

Research on the FMCG Supply Chain Quality Control Tower System and Operation Mechanism

The article used collaborative control theory, Internet of Things technology and the house of quality, and then proposed an innovative concept of FMCG supply chain quality control tower based on FMCG supply chain quality features. It elaborated its operation mechanism of the "five layers and three mechanisms" that provided theoretical support for quality control and traceability of FMCG.