On-demand Manufacturing Services Research Articles

Early adopters of Manufacturing-as-a-Service (MaaS): state-of-the-art and deployment models

PurposeCloud Manufacturing (CM) is the manufacturing version of Cloud Computing and aims to increase flexibility in the provision of manufacturing services. On-demand manufacturing services can be requested by users to the cloud and this enables the concept of Manufacturing-as-a-Service (MaaS). Given the considerable number of prototypes and proofs of concept addressed in literature, this work seeks real CM platforms to study them from a business perspective, in order to discover what MaaS concretely means today and how these platforms are operating.Design/methodology/approachSince the number of real applications of this paradigm is very limited (if the authors exclude prototypes), the research approach is qualitative. The paper presents a multiple-case analysis of 6 different platforms operating in the manufacturing field today. It is based on empirical data and inductively researches differences among them (e.g. stakeholders, operational flows, capabilities offered and scalability level).FindingsMaaS has come true in some contexts, and today it is following two different deployment models: open or closed to the provider side. The open architecture is inspired by a truly open platform which allows any company to be part of the pool of service providers, while the closed architecture is limited to a single service provider of the manufacturing services, as it happens in most cloud computing services.Originality/valueThe research shoots a picture of what MaaS offers today in term of capabilities, what are the deployment models and finally suggests a framework to assess different levels of development of MaaS platforms.

Sequence generation for multi-task scheduling in cloud manufacturing with deep reinforcement learning

Cloud manufacturing is a manufacturing model that aims to deliver on-demand manufacturing services to consumers. Scheduling is an important problem that needs to be addressed carefully and effectively for cloud manufacturing to achieve that aim. Cloud manufacturing allows consumers to submit their requirements to the cloud platform simultaneously and therefore requires cloud manufacturing scheduling systems to be able to handle multiple tasks effectively. It is further complicated when multiple composite tasks are submitted to the system and to be addressed. A vast majority of existing studies have proposed various algorithms, including meta-heuristics, heuristics, and reinforcement learning algorithms, to address cloud manufacturing scheduling (CMfg-Sch) problems, but only a very small fraction of them deal with scheduling of multiple composition tasks with deep reinforcement learning. In this work, we leverage DRL coupled with sequence generation for addressing CMfg-Sch problems. Different from all existing works, we first propose two sequence generation algorithms for generating scheduling sequences of multiple composite tasks prior to scheduling. Coupled with this a Deep Q-Networks (DQN) and a Double DQN-based scheduling algorithms are proposed, respectively. Performance of the proposed algorithms is compared against seven baseline algorithms using makespan, cost, and reliability as evaluation metrics. Comparison indicates that sequence generation algorithm II (SGA-II) overall has a greater advantage over algorithm I (SGA-I), especially in terms of the makespan, and the Double DQN-based scheduling algorithm outperforms the DQN-based algorithm, which in turn performs better than other baseline algorithms.

Cloud–edge collaboration task scheduling in cloud manufacturing: An attention-based deep reinforcement learning approach

Cloud Manufacturing (CMfg), as a service-oriented manufacturing mode, aims to provide consumers on-demand manufacturing services. The CMfg platform requires task scheduling technology to schedule manufacturing tasks efficiently, and improve resource utilization and customer satisfaction. Existing scheduling models for manufacturing tasks mainly consider maximizing the quality of service for customers but ignore the actual production execution, which will lead to low-quality execution or delayed delivery. To maximize customer satisfaction and balance production, this article studies a cloud–edge collaboration manufacturing task scheduling in CMfg (CETS). CETS refines manufacturing services deployed in the cloud to the factory process level, and schedules tasks according to the real-time production information on the edge side and manufacturing service information on the cloud side. Considering the dynamics of CETS and the complexity of state information in CETS, an attention-based deep reinforcement learning (DRL) algorithm is proposed to solve CETS. First, the CETS is mathematically represented and built as a partially observable Markov decision process. Second, on-policy maximum a posteriori policy optimization (V-MPO) with gated transformer-XL (GTrXL) named AV-MPO is developed. The effectiveness, training stability, generalizability, scalability, and robustness of AV-MPO are investigated. Rule-based algorithms and some state of art DRL algorithms, such as proximal policy optimization (PPO), soft actor-critic (SAC), and dueling deep q network (Dueling DQN), are compared with AV-MPO. The experimental results validate that AV-MPO can deal with the CETS problem more effectively.

An Efficient IIoT Gateway for Cloud–Edge Collaboration in Cloud Manufacturing

The cloud manufacturing system can provide consumers with on-demand manufacturing services, which significantly improve the utilization rate of distributed manufacturing resources and the response speed of personalized product needs. In the cloud manufacturing platform, the successful implementation of various industrial applications relies on the uploading and streaming of related field-level manufacturing data. For example, the realization of manufacturing service composition application should match the manufacturing tasks with distributed manufacturing resources according to their working state data and performance measurement data. Therefore, this paper proposes a data integration and analysis framework of a cloud manufacturing system based on cloud–edge collaboration and the Industrial Internet of Things (IIoT). A service-oriented information model is established to uniformly describe the related operational data and functional attributes of heterogeneous manufacturing resources. Secondly, a real-time transmission and integration method of high-volume operational field and sensor data based on message middleware is proposed to realize the remote monitoring of distributed manufacturing resources and efficient distribution of related data. Finally, a cloud–edge collaboration mechanism is put forward to train and update the parameters of various artificial intelligence models deployed at edge gateways. In the experiment, taking the computer numerical control (CNC) lathe as an example, the effectiveness of the proposed manufacturing resource access method is verified. Taking the fault diagnosis model of the CNC lathe as an example, the efficiency of the proposed cloud–edge collaboration mechanism for model updating is verified.

A privacy-preserving resource trading scheme for Cloud Manufacturing with edge-PLCs in IIoT

With the development of industrial Internet of things (IIoT), Cloud Manufacturing has been increasingly popular to the manufacturing industry. It can provide resource-sharing and on-demand manufacturing services as well as automatic collaborative manufacturing with the help of edge Programmable Logic Controllers (edge-PLCs). In such a system, there is a high risk of exposing user privacy and trading secret, due to exposure of sensitive transaction data to public servers. We propose a new privacy-preserving resource-trading scheme (PRTS), which leverages the concept of homomorphic cryptography and asymmetric searchable encryption, to simultaneously protect the privacy of the equipment factory and parts factories, while supporting best matching results in terms of parts parameters and price. Furthermore, a random forest-based method is applied to identify abnormal participants. The experimental results and security analysis show that the proposed scheme is accurate, effective, and secure, even under Off-line Keyword Guessing Attacks. Finally, encrypted data can resist analysis from mainstream machine learning techniques.

QoS-Aware Service Composition in Cloud Manufacturing: A Gale–Shapley Algorithm-Based Approach

Cloud manufacturing (CMfg) is an emerging paradigm that aims to provide on-demand manufacturing services over the Internet. Service composition as an important means for generating value-added services plays an important role in achieving the aim of CMfg. Most of previous works focused on exploring techniques of service composition for a single composite task using meta-heuristic algorithms. However, the issue of service composition for multiple composite tasks has rarely been considered. Meta-heuristic algorithms suffer from cumbersome parameter tuning as well as the tendency of getting into local optima. In addition, the effectiveness of different algorithms has not yet been fully explored when different degrees of constraints are imposed. Different from approaches in most of the previous works, this paper proposes an extended Gale–Shapley (GS) algorithm-based approach for service composition that allows generation of multiple service composition solutions effectively. Requirements with different constraints are considered. Experimental results indicate that: 1) meta-heuristic algorithms can be used in various scenarios with different degrees of constraints. However, they are incapable of finding the optimal solutions in situations with relatively loose constraints, and moreover, the failure rate of finding solutions for a batch of multiple tasks is high; 2) the dynamic programming (DP) is a method that is the most sensitive to constraints. It performs better only under loose constraints and in the case of a single requirement; and 3) the application range of the GS method proposed is wider than that of the DP method. It can achieve better performance when constraints are relaxed irrespective of task status (i.e., a single task or multiple tasks), and moreover, it can make more tasks find solutions in the multitask scenario without service reuse.

Logistics-involved QoS-aware service composition in cloud manufacturing with deep reinforcement learning

• Deep reinforcement learning is used to solve the QoS-aware service composition problem in cloud manufacturing with the consideration of logistics. • A DRL algorithm that combines the basic DQN algorithm, the dueling architecture, and the prioritized replay mechanism is proposed. • Experimental results demonstrate the effectiveness, high efficiency, adaptability, and scalability of the proposed DRL algorithm. Cloud manufacturing is a new manufacturing model that aims to provide on-demand manufacturing services to consumers over the Internet. Service composition is an essential issue as well as an important technique in cloud manufacturing (CMfg) that supports construction of larger-granularity, value-added services by combining a number of smaller-granularity services to satisfy consumers’ complex requirements. Meta-heuristics algorithms such as genetic algorithm, particle swarm optimization, and ant colony algorithm are frequently employed for addressing service composition issues in cloud manufacturing. These algorithms, however, require complex design flows and painstaking parameter tuning, and lack adaptability to dynamic environment. Deep reinforcement learning (DRL) provides an alternative approach for solving cloud manufacturing service composition (CMfg-SC) issues. DRL as model-free artificial intelligent methods enables a system to learn optimal service composition solutions through training, which can therefore circumvent the aforementioned problems with meta-heuristics algorithms. This paper is dedicated to exploring possible applications of DRL in CMfg-SC. A logistics-involved QoS-aware DRL-based CMfg-SC is proposed. A dueling Deep Q-Network (DQN) with prioritized replay named PD-DQN is designed as the DRL algorithm. Effectiveness, robustness, adaptability, and scalability of PD-DQN are investigated, and compared with that of the basic DQN and Q-learning. Experimental results indicate that PD-DQN is able to effectively address the CMfg-SC problem.

Logistics service scheduling with manufacturing provider selection in cloud manufacturing

Abstract In service-oriented manufacturing models, manufacturing resources in different enterprises are integrated and shared through network, cloud platforms, and logistics. On cloud manufacturing platforms, service providers offer on-demand manufacturing services to service demanders according to supply-demand matching results. As a special type of manufacturing services, logistics services provide transportation capabilities for production services and demanders. It is a critical issue to schedule logistics services efficiently, especially when manufacturer selections have been planned. This research focuses on the logistics scheduling problem in cloud manufacturing with pre-selected manufacturers. We analyze this optimization problem from aspects of tasks, production services, logistics services, and optimization objectives. Then a logistics scheduling method is proposed to reduce the average delivery time from manufacturers to customers. In the proposed method, the total time from start points of logistics to demanders is considered to reduce the average delivery time of all tasks. Based on four different scenarios, we build their scheduling models and run simulations to verify the effectiveness of the proposed method. Results show that the average task delivery time of the proposed method is shorter than three typical strategies.

Integrating Capacity and Logistics of Large Additive Manufacturing Networks

Among future manufacturing systems, we are going to see networks of intelligent and autonomous entities sharing manufacturing resources, knowledge and information. Possible advantages are relevant, such as increasing overall production efficiency and product variety as well as reducing responsiveness and lead times. This paper focuses on the architectures and dynamics of productive-demanding nodes in a Scattered Manufacturing (SM) Network, with an application in Additive Manufacturing scenarios. SM allows launching production orders everywhere anytime inside the domain of the network. These autonomous nodes can rely on on-demand manufacturing services by sharing resources in a geographically distributed network. One possible approach is the introduction of a platform to coordinate the dynamics along the network according to principles of sustainability, equated shared resources and transparency by managing communication activities among nodes. To identify variables/factors that affect the system, a unique model is proposed by combining different perspectives, which focus on: a) decomposition and localization of demanding node’s order into subtasks of variable size; b) tasks allocation criteria among geographically distributed nodes; c) logistics issues related to the localization of productive nodes. In particular, the model, with the aim of optimizing the overall manufacturing and logistics costs, suggests either logistics paths along the sub-network or tasks assignment criteria and scheduling in geographically distributed nodes.

A hybrid grey wolf optimizer algorithm with evolutionary operators for optimal QoS-aware service composition and optimal selection in cloud manufacturing

Cloud manufacturing (CMfg), as a new service-oriented technology, is aiming towards delivering on-demand manufacturing services over the internet by facilitating collaboration among different producers with distributed manufacturing resources and capabilities. To this end, addressing service composition and optimal selection (SCOS) problem has been the pivotal challenge. This NP-hard combinatorial problem deals with selecting and combining the available resources into a composite service to meet the user’s requirements while keeping up the optimal quality of service. This study proposes a new hybrid approach based on the recently developed grey wolf optimizer (GWO) algorithm and evolutionary operators of the genetic algorithm. The embedded crossover and mutation operators carry out a twofold functionality: (1) they make it possible to adapt the continuous structure of GWO to a combinatorial problem such as SCOS, and (2) they help to avoid the local optimal stagnation at the hunting process by providing more exploration strength. A series of experiments were designed and conducted to prove the effectiveness of the proposed algorithm, and the experimental results demonstrated that the proposed algorithm delivers superior performance compared with that of both existing discrete variations of GWO and genetic algorithm, especially in large-scale SCOS problems.

Cloud-based manufacturing equipment and big data analytics to enable on-demand manufacturing services

Abstract Making manufacturing as on-demand cloud services is a transformative paradigm to achieve the required business flexibility in the context of Industry 4.0 via enabling rapid configuration of loosely-connected manufacturing devices to develop highly customized products. The research in this paper aimed to fill the gap that there is a lack of a feasible solution for cloud-based manufacturing equipment that can provide on-demand manufacturing services accessible via the Internet. The technical challenges in developing cloud-based manufacturing equipment and the enabling technologies are discussed. A generic system architecture for cloud-based manufacturing equipment based on cyber-physical production systems and big data analytics is proposed, allowing manufacturing equipment to be connected to the cloud and made available for the provision of on-demand manufacturing services. An industry implementation in a world-leading machinery solution provider confirms that the proposed system architecture for cloud-based manufacturing equipment can successfully enable on-demand manufacturing services provisioned via the Internet and can be extended to businesses that endeavor to transform legacy production systems into cloud-based cyber-physical production systems.

Cloud Manufacturing Platform and Architecture Design

Abstract Cloud Manufacturing (CMfg) is an emerging networked manufacturing paradigm and service-oriented manufacturing model, in which distributed manufacturing resources are made available by providers according to consumers’ requirements as on-demand manufacturing services via networks (manufacturing clouds) and cloud manufacturing service platforms. Considering the concept of cloud manufacturing and its operation principle, the paper designs and introduces new proposal of cloud manufacturing service platform architecture, including the following six layers: physical resources layer, soft resources layer, virtual resources layer, services layer, application layer and service-oriented interface layer. After surveying majority of papers introducing architectures of CMfg service platforms, the paper recommends to map both hard as well as soft resources into cloud services from resource layers to virtual resource layer which highlights the core idea of the concept. The paper discusses also the fundamentals of the CMfg paradigm, introduces three groups of actors that participate in a cloud manufacturing system, as well as indicates briefly a typical hierarchy architecture of cloud manufacturing system.

Cloud Manufacturing

Cloud manufacturing is emerging as a new manufacturing paradigm which applies well-known basic concepts from cloud computing to manufacturing processes and deliver shared, ubiquitous, on-demand manufacturing services. It is an innovative, web-based manufacturing model. It is promising to transform today’s manufacturing industry from production-oriented to service-oriented, highly collaborative manufacturing of the future. This paper provides a brief introduction to cloud manufacturing.

Multi-agent-based scheduling in cloud manufacturing with dynamic task arrivals

Scheduling is a critical means for providing on-demand manufacturing services in cloud manufacturing. Multi-agent technologies provide an effective approach for addressing scheduling issues in cloud manufacturing, which, however, have rarely been used for solving the issue. This paper addresses scheduling issues in cloud manufacturing using multi-agent technologies. A multi-agent architecture for scheduling in cloud manufacturing is proposed firstly. Then, a corresponding multi-agent model is presented, which incorporates many-to-many negotiations based on an extended contract net protocol and takes into account dynamic task arrivals. Simulation results indicate the feasibility of the model and approach proposed.

An event-triggered dynamic scheduling method for randomly arriving tasks in cloud manufacturing

ABSTRACTCloud manufacturing (CMfg) is a service-oriented manufacturing mode and provides customers with on-demand manufacturing services via network. To solve the task scheduling problem in the dynamic CMfg environment with randomly arriving tasks, this article presents an event-triggered dynamic task scheduling (EDS) method. The event-triggered strategy is designed by considering arrival of new tasks and completion of first or middle subtasks in the subtask sequences to improve the timeliness of the service scheduler. The subtask-oriented strategy is applied to avoid service pre-emption so that the timely and effective task schedules can be obtained. The service time, logistic time and earliest available time of the candidate services are combined to choose the optimal services for triggered subtasks. Based on a case study of numerical control machining, the experimental results demonstrate that the average task execution time of the task schedules obtained through EDS was shorter than the other three task scheduling methods as tasks arrive randomly over time, and the total amount of occupied services during the scheduling period was not high.

Diverse task scheduling for individualized requirements in cloud manufacturing

ABSTRACTCloud manufacturing (CMfg) has emerged as a new manufacturing paradigm that provides ubiquitous, on-demand manufacturing services to customers through network and CMfg platforms. In CMfg system, task scheduling as an important means of finding suitable services for specific manufacturing tasks plays a key role in enhancing the system performance. Customers’ requirements in CMfg are highly individualized, which leads to diverse manufacturing tasks in terms of execution flows and users’ preferences. We focus on diverse manufacturing tasks and aim to address their scheduling issue in CMfg. First of all, a mathematical model of task scheduling is built based on analysis of the scheduling process in CMfg. To solve this scheduling problem, we propose a scheduling method aiming for diverse tasks, which enables each service demander to obtain desired manufacturing services. The candidate service sets are generated according to subtask directed graphs. An improved genetic algorithm is applied to searching for optimal task scheduling solutions. The effectiveness of the scheduling method proposed is verified by a case study with individualized customers’ requirements. The results indicate that the proposed task scheduling method is able to achieve better performance than some usual algorithms such as simulated annealing and pattern search.

The Dynamic Enterprise Network Composition Algorithm for Efficient Operation in Cloud Manufacturing

As a service oriented and networked model, cloud manufacturing (CM) has been proposed recently for solving a variety of manufacturing problems, including diverse requirements from customers. In CM, on-demand manufacturing services are provided by a temporary production network composed of several enterprises participating within an enterprise network. In other words, the production network is the main agent of production and a subset of an enterprise network. Therefore, it is essential to compose the enterprise network in a way that can respond to demands properly. A properly-composed enterprise network means the network can handle demands that arrive at the CM, with minimal costs, such as network composition and operation costs, such as participation contract costs, system maintenance costs, and so forth. Due to trade-offs among costs (e.g., contract cost and opportunity cost of production), it is a non-trivial problem to find the optimal network enterprise composition. In addition, this includes probabilistic constraints, such as forecasted demand. In this paper, we propose an algorithm, named the dynamic enterprise network composition algorithm (DENCA), based on a genetic algorithm to solve the enterprise network composition problem. A numerical simulation result is provided to demonstrate the performance of the proposed algorithm.