A semantic-level component-based scheduling method for customized manufacturing

doi:10.1016/j.rcim.2021.102144

Robotics and Computer-Integrated Manufacturing

Volume 71, October 2021, 102144

https://doi.org/10.1016/j.rcim.2021.102144 Get rights and content

Highlights

•
Information model obtains the job shop state through semantic reasoning.
•
Computation model generates the optimal schedules for guiding the control of manufacturing resources.
•
The integration of information model and computation model reduces the search of the solution space.

Abstract

To meet the increasingly complex needs of customers, scheduling faces challenges of the high uncertainty of product arrival in customized manufacturing (CM). This paper proposes a semantic-level component-based scheduling method to solve the uncertainty via the integration of the information model and the computation model. In our proposal, we first construct a component-based framework to illustrate the composition and execution mechanism of a component. Then we present a semantic-enriched information model to obtain the state of the shop floor through automatic semantic reasoning. Additionally, we build a computation model to abstract the stochastic scheduling process of CM. Finally, we design an iteration algorithm to solve the computation model through the interaction between the information model and computation model. In experiments, we show that for random arrivals of products, our proposal can ensure the timeliness of the learning and decision-making, and the task assignment performance is the best compared with the other two methods.

Introduction

In the past, mass production (MP) used the scale effect of production to reduce costs, improving the competitiveness of enterprises. However, times are changing, and the trends of manufacturing globalization, the diversification of consumer demands, and the shortening of product market cycles pose great challenges to traditional manufacturing enterprises [1]. Mass customized production (MCP) emerged to cope with this trend [2], which produces a variety of customized products using a mass production strategy to obtain benefits both from mass and customized production. However, to balance costs in MCP, manufacturers increase the types of products as much as possible to meet the diverse needs of customers, while manufacturers dominate the product design rather than customers in the design phase. Direct customer input to design will enable companies to increasingly produce customized products with shorter cycle-times and lower costs than those associated with MCP. Industry 4.0 provides the possibility to let customers participate wholeheartedly in the design phase [3]. Customized manufacturing (CM) [4] enabled by Industry 4.0 provides a competitive mode of producing a wide variety of batches and types of products with the direct needs of customers. Because of the new characteristics in CM, there are some new scheduling problems, for example, the frequency of changing production schedules usually cannot match the frequency of changing demands. This requires the shop floor to meet two scheduling prerequisites: 1. The shop floor can quickly obtain information about products and resources to agilely respond to computations to obtain scheduling results when new products arrive. 2. The shop floor has sufficient production capacity to process various products.

To support CM, advanced techniques and methods of Industry 4.0, such as industrial internet of things (IIoT) technology [5], service-oriented architectures (SOAs) [6], cloud manufacturing [7] and big data [8], must be adopted. Therefore, the cyber-physical production system (CPPS) concept [9] [10] has emerged to provide prospects to make traditional manufacturing move towards smart manufacturing, which adapts to the effects of product variability and manufacturing process disturbances. A CPPS integrates information processes (such as scheduling, simulation, execution, and monitoring) and physical manufacturing [11]. Albeit the existing CPPS architecture partially breaks the traditional automation pyramids, typical control and field levels activities still exist,leading to the prevention of peer-to-peer interaction between production tools and production elements [12]. The reason is that the asset data is structured in an ad-hoc manner for specific scenarios, and the data structures are usually inconsistent during different stages of the lifecycle. In conclusion, the sharing and reuse of information have not been resolved. Componentization [13] is an approach to breaking a system down into identifiable pieces that application developers independently deploy. This process is expressed as dividing things into different functional modules based on their different tasks and encapsulating each function into individual components. In addition, the components are independent of each other. Thus, the componentization of manufacturing elements is a promising way to decouple various production elements and increase the reusability of information to achieve a true fusion of the cyber and the physical realms.

The remainder of this paper is organized as follows. Section 2 reviews the literature. Section 3 describes the concept of CPPS componentized scheduling. Section 4 presents the information model. Section 5 constructs the computation model. Section 6 discusses integrating the information and computation models and proposes an optimized algorithm. Section 7 presents a real-world case. Section 8 discusses the experiment results. Finally, Section 9 summarizes the work and future research directions.

Section snippets

Literature review

In CM, products are released and arrive at a shop over time. The release date, the processing time of products and machine breakdowns are stochastic and are not known in advance. It is difficult and sometimes impossible to compute optimal schedules. To build suitable dynamic scheduling systems, several methods have been proposed thus far.

Some studies have focused on the rescheduling method based on AI techniques such as neural networks (NNs), expert systems (ESs), fuzzy logic, simulated

Framework of CPPS components

In CM, the challenge with respect to scheduling is how to organize resources to meet the requirements of the time limit and the production cost. When new products or new resources are introduced, the shop floor must be able to detect and promptly respond to changes in the production environment. To achieve this goal, it is necessary to componentize assets into informational abstractions, where the assets represent the entities that can affect production. In shop floor design and operation,

Information model

For the phases of scheduling, only three kinds of CPPS components are considered: products, processes, and resources [31] [32]. These three types of CPPS components should be classified with suitable granularity to ensure adequate flexibility. In the same type of CPPS components, several fine-granularity components are compounded into composite components. Thus, the product-process-resource (PPR) hierarchies are presented in Fig. 2.

Resources provide processing capabilities and participate in

Computation model

In CM, when a machine becomes idle and several jobs are waiting in front of it, we will decide which job should be processed next. There are many types of resources, and their performances vary greatly. In addition, the process time and quantity of products often change. The traditional scheduling method based on PPR does not fully decouple the PPR concept at the semantic information level, and there is still partial coupling between production demand and resource capacity. Once the scheduling

Integration of the information model and the computation model

In the computation model, the optimal task allocation policy enables the shop floor to obtain the maximum comprehensive profit. This means that all resources can achieve the best task allocation within a limited time. Because of the stochastic characteristics of product arrivals, there is no fixed action for task allocation. Each task allocation affects the comprehensive profit of the shop floor, and the number of possible solutions to the optimal policy set is enormous. To meet the limited

Case study

This paper employs an intelligent shop floor as a test platform for functional verification. The test platform widely applies the key technologies developed in this paper. Two experiments are conducted to verify the query performance of the information model and to verify the task assignment performance of the computation model. The hardware layout, shown in Fig. 9, includes a server powered by a 3.5 GHz CPU as a private cloud, two computer numerical control (CNC) stations, a laser labeling

Results and discussion

Fig. 10 shows the results of experiment 1, indicating the relationship between the resource quantity and the average response time. With the increase in the feature quantity, the average response time tends to gradually rise. Moreover, the time spent matching a single feature increases with the increase in the number of resources. The reason is that when matching more features or more resources, the memory consumption of the information model instances increases, resulting in an increase in the

Conclusion

This paper proposes a semantic-level component-based scheduling method to implement a shop floor with the uncertainty of product arrival. We formulate the design goal as figuring the maximized production profit. Toward this direction, we architect a component-based framework that describes the composition and execution mechanism of a component. On this basis, we put forward a semantic-enriched information model to decouple the PPR concept and obtain the state of shop floor through automatic

Author statement

Under supervision byDi Li, Hao Tangperformed experimentpreparation and data analysis. Hao Tang builtthe information model and computation model. Di Li performed calculations and performed experiment. All authors read and contributed to the manuscript.

Declaration of Competing Interest

The authors declared that they have no conflicts of interest to this work.

We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted

References (40)

R. Zhong et al.
RFID-enabled real-time manufacturing execution system for mass-customization production
Robot. Comput. Integr. Manuf.
(2013)
L. Monostori
Cyber-physical production systems: roots, expectations and R&D challenges
Procedia Cirp
(2014)
L. Monostori et al.
Cyber-physical systems in manufacturing
Cirp Annals
(2016)
L. Wang et al.
Current status and advancement of cyber-physical systems in manufacturing
J. Manuf. Syst.
(2015)
A. Gharehgozli et al.
A fuzzy-mixed-integer goal programming model for a parallel-machine scheduling problem with sequence-dependent setup times and release dates
Robot. Comput. Integr. Manuf.
(2009)
M. Dai et al.
Energy-efficient scheduling for a flexible flow shop using an improved genetic-simulated annealing algorithm
Robot. Comput. Integr. Manuf.
(2013)
N. Kundakcı et al.
Hybrid genetic algorithms for minimizing makespan in dynamic job shop scheduling problem
Comp. Indust. Eng.
(2016)
D. Šormaz et al.
SIMPM–Upper-level ontology for manufacturing process plan network generation
Robot. Comput. Integr. Manuf.
(2019)
J. Liang
An ontology-oriented knowledge methodology for process planning in additive layer manufacturing
Robot. Comput. Integr. Manuf.
(2018)
P. Maropoulos et al.
Resource-aware aggregate planning for the distributed manufacturing enterprise
CIRP Annals
(2002)

E. Sirin et al.

Pellet: a practical owl-dl reasoned

J. Web Semant.

(2007)

H. Tang et al.

CASOA: an architecture for agent-based manufacturing system in the context of Industry 4.0

IEEE Access

(2018)

T. Wang et al.

Industry 4.0: a way from mass customization to mass personalization production

Adv. Manuf.

(2017)

J. Wan et al.

Artificial-Intelligence-Driven Customized Manufacturing Factory: key Technologies, Applications, and Challenges

Proc. IEEE

(2020)

H. Tang et al.

A Reconfigurable Method for Intelligent Manufacturing Based on Industrial Cloud and Edge Intelligence

IEEE Internet of Things J.

(2019)

J. Morgan et al.

Enabling a ubiquitous and cloud manufacturing foundation with field-level service-oriented architecture

Int. J. Comp. Integrat. Manuf.

(2017)

C. Zhao et al.

Modeling of service agents for simulation in cloud manufacturing

Robot. Comput. Integr. Manuf.

(2020)

D. Li et al.

A big data enabled load-balancing control for smart manufacturing of Industry 4.0

Cluster Comput.

(2017)

S. Karnouskos et al.

Key Directions for Industrial Agent Based Cyber-Physical Production Systems

C. Bock

Componentization in the Systems Modeling Language

Syst. Eng.

(2014)

Cited by (11)

An Expandable and Generalized Method for Equipment Information Reflection in Digital Twin Workshop Systems
2024, Robotics and Computer-Integrated Manufacturing
A production workshop is equipped with diverse equipment and machinery, and manufacturing data generated by different equipment show certain differences. Therefore, developing a method with an information presentation capability for a digital twin workshop system (DTWS) can be challenging due to the numerous information sources and types. The level of detail provided for a DTWS is directly related to the management comprehension of the production status of a physical workshop. Thus, arranging and presenting different data effectively represents a significant challenge. To address this challenge, this study proposes an equipment information self-reflection method with expandability for DTWSs. First, a generic equipment information model with an expandability is developed. This model encompasses all relevant data types in a workshop. The analysis of these data is conducted from two perspectives: workshop equipment categorization and equipment internal composition. The analysis results are used for the generation of extensible styles for property elements and function components. For the DTWSs, these two perspectives offer a standardized specification for equipment information presentation. In addition, a self-reflection method is introduced to address the demand for generating diverse information types ranging from equipment-based to extensible types. The system framework and operational flow of this method are explained in detail. To verify the proposed method, this paper conducts a practical case study using the DTWorks software development kit to simulate a real workshop. The proposed self-reflection method is implemented into the DTWorks. The disparities in equipment information types are analyzed for both pre- and post-implementation of the proposed method. The obtained results demonstrate the efficiency and advantages of the proposed self-reflection method. Finally, the results show that the proposed method exhibits advantageous performance regarding intricate business contexts.
Knowledge graph modeling method for product manufacturing process based on human–cyber–physical fusion
2023, Advanced Engineering Informatics
The data generated in the product manufacturing process are usually distributed in different formats, triggering fragmented knowledge and disconnected information. To address this problem, we present a knowledge graph modeling method for the product manufacturing process. First, the concepts of human–cyber–physical (HCP) elements are analyzed in detail. The HCP-related classes, attributes, and relations are defined in a formalized manner in the ontology modeling process. Second, a knowledge graph model for the product manufacturing process (KGM/PMP) is constructed by three steps, including knowledge extraction, knowledge fusion, and knowledge reasoning. When constructing the KGM/PMP model, a deep learning method called BERT-D’BiGRU-CRF is presented to automatically extract knowledge from the manufacturing data. Moreover, a set of reasoning rules are designed to infer new knowledge. Finally, a case study is carried out to validate the effectiveness of the proposed method. The validity of the BERT-D’BiGRU-CRF method on knowledge extraction is verified by comparing performance with four other methods. The applicability of the knowledge graph model is demonstrated through developing a prototype system. With this system, manufacturing knowledge can be provided for the demanders rapidly and accurately.
Solving job scheduling problems in a resource preemption environment with multi-agent reinforcement learning
2022, Robotics and Computer-Integrated Manufacturing
Citation Excerpt :
The Job Shop Scheduling Problem (JSSP), serving as an abstraction of the manufacturing environment such as the cloud manufacturing [1,2] and the smart manufacturing system [3,4], has been comprehensively studied in the past decades.
In smart manufacturing, robots gradually replace traditional machines as new processing units, which have significantly liberated laborers and reduced manufacturing expenditure. However, manufacturing resources are usually limited so that the preemption relationship exists among robots. Under this circumstance, job scheduling puts forward higher requirements on accuracy and generalization. To this end, this paper proposes a scheduling algorithm to solve job scheduling problems in a resource preemption environment with multi-agent reinforcement learning. The resource preemption environment is modeled as a decentralized partially observable Markov decision process, where each job is regarded as an intelligent agent that chooses an available robot according to its current partial observation. Based on this modeling, a multi-agent scheduling architecture is constructed to handle the high-dimension action space issue caused by multi-task simultaneous scheduling. Besides, multi-agent reinforcement learning is employed to learn both the decision-making policy of each agent and the cooperation between job agents. This paper is novel in addressing the scheduling problem in a resource preemption environment and solving the job shop scheduling problem with multi-agent reinforcement learning. The experiments of the case study indicate that our proposed method outperforms the traditional rule-based methods and the distributed-agent reinforcement learning method in total makespan, training stability, and model generalization.
KSPMI: A Knowledge-based System for Predictive Maintenance in Industry 4.0
2022, Robotics and Computer-Integrated Manufacturing
Citation Excerpt :
However, this approach only applies data-driven methods for predict degradation, which lacks the incorporation of domain knowledge about system’s failure mechanism (e.g. friction, tear, wear and crack growth). In [46], a semantic-enriched information model is developed to assess the state of shop floor by semantic reasoning. However, in the semantic information model, there lacks a mechanism for evaluating the quality of rules.
In the context of Industry 4.0, smart factories use advanced sensing and data analytic technologies to understand and monitor the manufacturing processes. To enhance production efficiency and reliability, statistical Artificial Intelligence (AI) technologies such as machine learning and data mining are used to detect and predict potential anomalies within manufacturing processes. However, due to the heterogeneous nature of industrial data, sometimes the knowledge extracted from industrial data is presented in a complex structure. This brings the semantic gap issue which stands for the lack of interoperability among different manufacturing systems. Furthermore, as the Cyber-Physical Systems (CPS) are becoming more knowledge-intensive, uniform knowledge representation of physical resources and real-time reasoning capabilities for analytic tasks are needed to automate the decision-making processes for these systems. These requirements highlight the potential of using symbolic AI for predictive maintenance.
To automate and facilitate predictive analytics in Industry 4.0, in this paper, we present a novel Knowledge-based System for Predictive Maintenance in Industry 4.0 (KSPMI). KSPMI is developed based on a novel hybrid approach that leverages both statistical and symbolic AI technologies. The hybrid approach involves using statistical AI technologies such as machine learning and chronicle mining (a special type of sequential pattern mining approach) to extract machine degradation models from industrial data. On the other hand, symbolic AI technologies, especially domain ontologies and logic rules, will use the extracted chronicle patterns to query and reason on system input data with rich domain and contextual knowledge. This hybrid approach uses Semantic Web Rule Language (SWRL) rules generated from chronicle patterns together with domain ontologies to perform ontology reasoning, which enables the automatic detection of machinery anomalies and the prediction of future events’ occurrence. KSPMI is evaluated and tested on both real-world and synthetic data sets.
Knowledge discovery using an enhanced latent Dirichlet allocation-based clustering method for solving on-site assembly problems
2022, Robotics and Computer-Integrated Manufacturing
Citation Excerpt :
Similarly, Wang et al. [8] used bi-level feature extraction-based text mining to detect anomalies and improve the fault diagnostics of railway systems. Li and Tang [9] proposed a semantic-based model to sense the shop floor state through automatic semantic reasoning for solving uncertainty of product demands arrivals in customized manufacturing. Other text-mining methods are based on semantic ontological models, the semantic web, and knowledge graphs.
Prompt responses to problems/faults arising in an assembly workshop are crucial in terms of production reliability and efficiency. However, human-dependent tasks are time-consuming and prone to error. In this paper, we propose a knowledge discovery approach. We extract the patterns of associations between texts in problem-solving records to generate appropriate solutions automatically. First, we use an enhanced latent Dirichlet allocation (EnLDA) technique to explore the document-topic and topic-word distributions of a text corpus recording assembly problems, causes, and solutions. To increase accuracy, we adjust the elements of the document-term matrix, and we assign term frequency-inverse document frequencies. Second, we use the Refining Density-based Spatial Clustering of Application with Noise (Rf-DBSCAN) algorithm for text clustering. This refines the distances among topic distribution vectors and incorporates noise objects into clustering. This clusters textual documents with similar semantic information, maximizing information retention. Third, we use the Apriori algorithm to identify pattern associations among document clusters that represent the problems, causes, and solutions. We perform a case study using field data from an automobile assembly workshop. The results show that the method retrieves hidden but valuable information from textual records. The decision support knowledge facilitates assembly problem-solving.
Toward smart manufacturing scheduling from an ontological approach of job-shop uncertainty sources
2022, IFAC-PapersOnLine
An integral application of the enabling technologies of Industry 4.0 in the job-shop scheduling problem (JSSP) must contemplate the automation and autonomy of the involved decision-making processes as a goal, which is the main purpose of the smart manufacturing scheduling (SMS) paradigm. In a real production context, uncertainty acts as a barrier that hinders this goal being met and, therefore, any SMS model should integrate uncertainty generators in one way or another. This paper proposes an ontological framework that identifies and structures the entities shaping the joint domain formed by the job-shop scheduling process in its itinerary toward the SMS paradigm, the sources of uncertainty that it faces, and the interrelationship type that link these entities. This ontological framework will serve in future research as a conceptual basis to design new quantitative models that, from a holistic perspective, will address the stochasticity of manufacturing environments and incorporate the management of disturbances into the realtime resolution of automatic and autonomous job-shop scheduling.

View all citing articles on Scopus

View full text

Full length ArticleA semantic-level component-based scheduling method for customized manufacturing

Highlights

Abstract

Introduction

Section snippets

Literature review

Framework of CPPS components

Information model

Computation model

Integration of the information model and the computation model

Case study

Results and discussion

Conclusion

Author statement

Declaration of Competing Interest

Robot. Comput. Integr. Manuf.

Procedia Cirp

Cirp Annals

J. Manuf. Syst.

Robot. Comput. Integr. Manuf.

Robot. Comput. Integr. Manuf.

Comp. Indust. Eng.

Robot. Comput. Integr. Manuf.

Robot. Comput. Integr. Manuf.

CIRP Annals

J. Web Semant.

CASOA: an architecture for agent-based manufacturing system in the context of Industry 4.0

IEEE Access

Industry 4.0: a way from mass customization to mass personalization production

Adv. Manuf.

Artificial-Intelligence-Driven Customized Manufacturing Factory: key Technologies, Applications, and Challenges

Proc. IEEE

A Reconfigurable Method for Intelligent Manufacturing Based on Industrial Cloud and Edge Intelligence

IEEE Internet of Things J.

Enabling a ubiquitous and cloud manufacturing foundation with field-level service-oriented architecture

Int. J. Comp. Integrat. Manuf.

Modeling of service agents for simulation in cloud manufacturing

Robot. Comput. Integr. Manuf.

A big data enabled load-balancing control for smart manufacturing of Industry 4.0

Cluster Comput.

Key Directions for Industrial Agent Based Cyber-Physical Production Systems

Componentization in the Systems Modeling Language

Syst. Eng.

Full length Article
A semantic-level component-based scheduling method for customized manufacturing