TheComputing Power Network(CPN) is a distributed system that integrates computing resources to optimize utilization, but ensuringQuality of Service(QoS) is challenging due to high demand and complex heterogeneous connections. Accurate workload prediction is essential for maintaining QoS, yet the diverse and complex user requirements in CPN make prediction difficult. To address this challenge, we propose an ensemble workload prediction model with fluctuation division control for workload prediction in CPN, comprising three key components. First, we use theThree-Way Decision(3WD) approach to partition workload fluctuations, controlling granularity thickness and applying clustering to capture dynamic workload characteristics. Second, we develop tailored prediction methods for each of the three partitioned regions and ensembles them to enhance overall prediction performance. Third, the ensemble prediction method is applied to each region to obtain the final predicted values. The proposed method introduces an innovative fluctuation division control strategy for characteristic mining to capture dynamic workload fluctuation patterns and designs the effective ensemble workload prediction model deal with the problem of non-stationary workload prediction in CPN. Experimental results on trace datasets from Alibaba and Dinda demonstrate that the proposed model improves the higher average prediction accuracy by up to 26.06%$\sim$66.4% than the comparison methods.

Resource failures are commonly viewed in automated manufacturing systems (AMSs), such as control errors and mechanical failures. Therefore, a robustness analysis and control policy are highly desirable. Different methods are proposed by researchers and practitioners. However, they are incapable of performing robustness analysis for partially observable AMSs since few works consider the existence of unobservable events. In this article, we solve the robustness analysis and control problem for partially observed AMSs in the paradigm of Petri nets (PNs). A marking is denoted as a robust (resp., nonrobust) one if from it the system can (resp., cannot) proceed continuously even though there exist resource failures. We show that the robustness of markings under partial observation can be determined using a compact reachability graph (RG), namely, a remarkably reduced RG. Then, we define robustness observability, and propose a simple method to check the robustness observability with respect to a given set of robust markings. Finally, we design robust controllers under partial observation so as to ensure continuous production of partially observable AMSs against resource failures.

Collaborative perception, fusing information from multiple agents, can extend perception range so as to improve perception performance. However, temporal asynchrony in real-world environments, caused by communication delays, clock misalignment, or sampling configuration differences, can lead to information mismatches. If this is not well handled, then the collaborative performance is patchy, and what's worse safety accidents may occur. To tackle this challenge, we propose CoDynTrust, an uncertainty-encoded asynchronous fusion perception framework that is robust to the information mismatches caused by temporal asynchrony. CoDynTrust generates dynamic feature trust modulus (DFTM) for each region of interest by modeling aleatoric and epistemic uncertainty as well as selectively suppressing or retaining single-vehicle features, thereby mitigating information mismatches. We then design a multi-scale fusion module to handle multi-scale feature maps processed by DFTM. Compared to existing works that also consider asynchronous collaborative perception, CoDynTrust combats various low-quality information in temporally asynchronous scenarios and allows uncertainty to be propagated to downstream tasks such as planning and control. Experimental results demonstrate that CoDynTrust significantly reduces performance degradation caused by temporal asynchrony across multiple datasets, achieving state-of-the-art detection performance even with temporal asynchrony. The code is available at https://github.com/CrazyShout/CoDynTrust.

User delegation is a type of access control mechanisms that allow a person to delegate all or part of his$/$her authorities to others. It is an important access control policy that can significantly improve workflow flexibility. A crucial requirement after delegations is to check the satisfiability of workflow, i.e., determining whether there exists a valid execution schedule ($ES$) in the workflow that can ensure it to proceed from the beginning to the end while satisfying all authorization constraints. Existing works perform in a centralized manner to find such a valid$ES$by exploring all$ES$s. Unfortunately, the enumeration of$ES$s is highly inefficient from the perspective of computational complexity. This study proposes a novel approach to overcoming this difficulty in a distributed way. First, we use Petri nets (PNs) to formalize workflow and user delegations. Then, we directly specify a conflict-free workflow that satisfies all authorization constraints. Hence, determining the satisfiability of a workflow amounts to directly checking the existence of$ES$s in the corresponding conflict-free workflow. Finally, we present a distributed strategy, which is of polynomial complexity, to determine the existence of$ES$s.

The problem considered in this paper is the verification and enforcement of separation of duty (SoD) constraints in attribute based access control (ABAC) systems. We propose an efficient algorithm for checking the satisfiability of SoD constraints. It is based on the idea of partitioning all permissions of SoD constraints into two classes so as to compute the minimal number of users to accomplish each class of permissions, respectively. As a result, several SoD constraints with certain class of permissions can be verified in polynomial time. Experimental results show that our method performs well compared with existing ones. When SoD violations occur, a 0-1 integer programming (IP) based enforcement solution is presented such that SoD violations can be solved once for all and it is provably shown that such solution does not result in the violation of other SoD constraints.

Given a role-based access control (RBAC), resiliency checking problem (RCP) aims at determining whether every permission is executed by a user and all authorization constraints are satisfied when some users become absent. Although the problem is computationally hard, desirable solutions are still expected so as to guarantee the continuity of access control. In this article, we solve RCP for RBAC based on constraint enforcement and mathematical programming. We use Petri nets (PNs) to formalize RBAC. It is shown that each separation of duty constraint imposed on a PN modeling of RBAC can be enforced by a maximally permissive PN-based control structure. After implementing such control structure on the PN modeling of RBAC, we can obtain an admissible RBAC. We show that RCP of RBAC can be transformed into another problem, which determines whether each permission can be executed by a user in the admissible RBAC against the absence of some users. An integer linear programming-based approach is presented to accomplish such verification. The comparison between our approach and the existing one is given to illustrate the effectiveness and efficiency of ours.

There are two foundational problems in automated manufacturing systems. One is to determine their robustness (i.e., checking whether a marking is robust or nonrobust) while the other is to determine their liveness (i.e., determining whether a marking is live, bad, deadlock, or livelock). However, existing methods deal with them separately. This renders the existing methods inefficient in practice. In this article, we investigate the relation between robustness and liveness. First, we show how to define robustness in different net systems, i.e., the live, bounded, and nonreversible or reversible net systems. Second, we present a reachability graph-based method to assess the robustness of markings. Third, we clarify the relation between robustness and liveness, and conclude that liveness is a special case of robustness, under which the set of unreliable transitions is null. As a result, the robustness determination method developed in this article proves to be much general and can be used to check the liveness of each marking.

Deadlock avoidance of flexible manufacturing systems (FMSs) has received increasing attention from both academic and industrial communities. There have been a large number of different types of deadlock avoidance policies discussed in the literature. However, how to avoid deadlocks in an efficient way is still one of the major obstacles, especially for large systems. In this paper, we propose a new Petri net structure, i.e., event circuit structures ($ESs$), based technique to overcome this difficulty. First, we provide details of$ESs$and develop an algorithm to calculate$ESs$in the systems of sequential systems with shared resources ($S^{4}Rs$). Second, we analyze the liveness of$S^{4}Rs$using undermarked$ESs$. A necessary and sufficient condition between undermarked$ESs$and deadlocks of$S^{4}Rs$is established. Third, we describe how undermarked$ESs$can be applied to avoid deadlocks for$S^{4}Rs$. Only structure information is needed during this procedure, thereby improving the efficiency and convenience of deadlock avoidance. Several examples are presented to illustrate our approach. Note to Practitioners—Deadlock avoidance of flexible manufacturing systems (FMSs) is extremely important in real-world manufacturing scenarios. A large body of deadlock avoidance policies are presented in the existing literature. Through an effective deadlock avoidance policy, all deadlocks can be prevented from happening in advance, so as to avoid the reallocation of resources and the re-execution of deadlocked processes. This shortens the production cycle of systems and improves the utilization of resources. However, most existing approaches suffer from formidable computational difficulty since they necessarily rely on the whole reachability graph to avoid deadlocks. In this paper, we present event circuit structures ($ESs$) as a new technique for deadlock avoidance. We show that undermarked$ESs$can be used to avoid deadlocks by using only key structure information instead of complicated state information. Thus, it not only can greatly improve the efficiency of predicting deadlock markings for FMSs, but also reduce operating costs as much as possible while ensuring stable operation of FMSs.

In this paper, we address the robustness analysis problem of automated manufacturing systems with uncontrollable events in the paradigm of Petri nets (PNs). First, we formalize unreliable resource failures as the removal of all ingoing transitions of unreliable resource places (denoted by unreliable transitions hereafter). Second, we obtain a necessary and sufficient condition to check the robustness of markings, so called robustness controllability theorem (abbreviated as RCT hereafter) in the paradigm of reduced reachability graph (abbreviated as R2G hereafter). All markings involved in the R2G of a PN are equivalent to that of its reachability graph, except that all arcs associated to unreliable transitions are removed from R2G. Based on RCT, the robustness of all markings in an R2G can be determined. An example is proposed to illustrate the approach. Note to Practitioners—In reality, it is an urgent issue to analyze the behaviors of automated manufacturing systems (AMSs) so as to guarantee their stable operation against resource failures, e.g., the missing of a signal or the failure of a sensor. In this connection, different methods are proposed to deal with the robustness analysis and control problem of AMSs with unreliable resources. The objective is to avoid any deadlock in the AMSs or to ensure the liveness of the subsystems that require only reliable resources when there exist resource failures. Due to limited actuating and sensing abilities, AMSs may be partially controlled, i.e., there exist events whose firing may not be inhibited by an external action. However, fewer research works consider this practical situation when handling robustness analysis and control issue, which renders the existing approaches impracticable. In this paper, we solve the robustness analysis problem of AMSs with uncontrollable events by using Petri nets. A necessary and sufficient condition is proposed to check the robustness of markings, called robustness controllability theorem (abbreviated as RCT hereafter) in the paradigm of reduced reachability graph. With the aid of RCT, the robustness of all markings can be determined so as to guarantee the flexibility of AMSs.

This paper focuses on constraint verification and violation resolution for Petri nets (PNs) modeling of role-based access control (RBAC) policy. Checking the satisfiability of authorization constraints imposes a major challenge when the number of states of a target system is large. To overcome this difficulty, we provide three necessary and sufficient conditions to check three different constraints, namely Separation of Duties (SoDs), Binding of Duties (BoDs), and Constraints of Cardinality (CoCs). The proposed results are based on the solutions of integer linear programming problems (ILPs). By relying on an ILP formulation that does not require the explicit computation of the net reachability set, the proposed approach is particularly well suited for large-size PNs. When the given system does not satisfy a considered constraint, the objective is to propose a suitable violation resolution strategy to correctly enforce the given constraint. In this paper, enforcement of control places and administration of RBAC are presented to solve the SoD, BoD, and CoC violations. All violations can be corrected in a once for all manner while simultaneously ensuring the satisfaction of all other constraints. The comparison between our approach and the existing ones is given to illustrate the effectiveness and efficiency of ours.

For real-world automated manufacturing systems (AMSs), the failures of multiple unreliable resources are very common. Maximal permissiveness is a key aspect for evaluating the performance of robust control strategies of failure-prone AMSs. However, there exist fewer works in the existing literature that can achieve maximally permissive robustness analysis. This may impede the applicability of existing robust control strategies in realistic AMSs. This article tackles the issue in the paradigm of Petri nets (PNs). We present an innovative mechanism to formalize unreliable resource failures by preventing all ingoing transitions of unreliable resource places from firing. Both reachability graph-based and integer linear programming-based methods are proposed to analyze the robustness of AMSs with either single or multiple unreliable resources. Several examples are proposed to illustrate the approach.

This paper studies the maximally permissive robustness analysis of automated manufacturing systems (AMSs) with unreliable resources in the paradigm of Petri nets (PNs). Two types of robust markings, i.e., strongly robust markings and weakly robust markings, are defined in this paper. We propose robustness equivalence and non-robustness equivalence to characterize the markings that exhibit the same robustness and non-robustness, respectively. Reachability graph (or RG hereafter) is directly used to determine the robustness of markings; however, it is difficult to use in large-scale systems due to formidable computational difficulty. As an alternative, we present a reduced reachability graph (or R2G hereafter) based necessary and sufficient condition to check the robustness of markings, in terms of the liveness analysis of markings in R2G. We show that all safe markings of an R2G correspond to strongly robust markings of the corresponding RG, and deadlock markings as well as their bad markings and livelock markings as well as their bad markings of an R2G correspond to non-robust markings and weakly robust markings of the corresponding RG, respectively. Hence, the robustness of markings in an RG can be determined effectively and efficiently through the liveness analysis of markings in the corresponding R2G. Note to Practitioners—In practical manufacturing scenarios, it is urgent to analyze and control the automated manufacturing systems (AMSs) so as to ensure their continual production against any resource failure. If the failures of an AMS are not handled gracefully, the whole system may fall into a blocking. As a consequence, system production does not meet rapid manufacturing goals and objectives. In this paper, we focus on the maximally permissive robustness analysis of AMSs with unreliable resources in the paradigm of Petri nets. We define two types of robustness in terms of markings, i.e., strong robustness and weak robustness. From the viewpoint of reachability graph, we propose robustness equivalence and non-robustness equivalence among markings and present the procedures to check the robustness of markings. Furthermore, a set of necessary and sufficient conditions are established to provably ensure that the robustness of markings can be determined through liveness analysis in a reduced reachability graph. Therefore, the robustness of markings can be determined in a computationally efficient way.

The problem under consideration in this paper is how to avoid deadlocks and livelocks in the paradigm of Petri nets (PNs). Although deadlock and livelock avoidance has been extensively studied in existing literature, fewer results can be applied to practical automated manufacturing systems (AMSs) due to expensive computations and reduced permissiveness. In this paper, we propose an efficient and maximally permissive control scheme to avoid deadlocks and livelocks by using critical distance. From the perspective of the reachability graph of a PN, critical distance is equal to the maximum number of transitions among all transition sequences from any critical state to its corresponding deadlock or livelock state. First, we show how to calculate the critical distance of a PN in an efficient way by using a portion of reachable states. Then, local reachability graphs are established based on critical distance to avoid deadlocks and livelocks. The provided policy is maximally permissive since only the necessary minimum number of illegal transitions are forbidden at each state. Several representative examples are presented to illustrate our approach. Note to Practitioners—Deadlock and livelock avoidance is a crucial problem in automated manufacturing systems (AMSs). Various methods have been proposed to deal with deadlocks and livelocks by researchers and practitioners. However, there exist fewer works that focus on deep exploration regarding how long a deadlock or livelock will occur from a safe state. As a consequence, the existing works suffer from the disadvantages of being too aggressive or conservative, i.e., checking the whole state space or forbidding many acceptable states when avoiding deadlocks and livelocks. Through the analysis of a partial of reachable states, this paper derives the critical distance of Petri nets modeling AMSs, which is equal to the maximum number of transitions among all transition sequences from any critical marking to its corresponding deadlock or livelock marking. All deadlocks and livelocks, if exist, can be detected at any given marking within the critical distance. Our approach not only is maximally permissive but also explores only a partial of states, thereby mitigating state explosion problem.

Much attention is being paid to security requirements of workflow processes with authorization policies, e.g., safety properties, liveness properties, separation of duties, binding of duties, and constraints of cardinality. However, existing methods neglect the execution condition of activities and the logical structures among activities along with their time attributes, suffer from low efficiency when checking the security requirements of large-scale and structurally complex workflow processes, and provide no solutions as a response to the violations of various security requirements. Thus, existing methods cannot guarantee the absolute security and smooth execution of such workflow processes. In this article, we propose a security team timed automaton (STTA) based approach to analyzing security requirements in timed workflow processes. First, we construct STTAs for timed workflow processes with authorization policies. Second, security requirements are automatically verified based on STTAs. Third, based on two effective strategies, we provide solutions to violated security requirements, if any. Compared with the existing methods, our approach can not only formally describe and analyze five commonly-viewed and frequently-adopted security requirements for timed workflow processes and dramatically decrease their temporal and spatial complexity for verification, but also provide solutions to the violations of security requirements so as to implement the security management of workflow processes.

Separations of Duties (SoDs) are an important class of security requirements in business process management. Their violation may result in system misuse and fraud, leading to economic losses or legal implications. Hence, it is of paramount importance to ensure that a business process meets all SoDs. Existing works usually adopt model checking to verify SoDs. However, building formal models that simultaneously account for both workflow and SoDs is a time-consuming and error-prone activity. In this article, we propose a new approach to specifying and enforcing SoDs in business processes using Petri nets (PNs). First, we derive a necessary and sufficient condition for the SoD violations from the viewpoint of structure and marking of PNs. We show that the SoD constraints can be enforced by disallowing the process to reach certain markings, with the constraints being written as linear inequalities. Then, we design supervisors to enforce SoDs in an off-line and a real-time manner, respectively, based on the linear inequalities. Meanwhile, inequality analysis is provided for the structural simplicity of supervisors. Finally, the complexity analysis of our approach and the comparison with the work in the literature are given to illustrate the effectiveness and efficiency of ours.

Generalized mutual exclusion constraints (GMECs) are a class of state specifications on Petri nets (PNs). They are generally enforced on the nets by a simple control structure called control places (monitors). Unfortunately, this conventional procedure is implemented in an offline and monolithic manner, which suffers from computational difficulty and is arduous for the control of a system in real time. Additionally, the flexibility and fault tolerance of such a method are somewhat unacceptable, and the method suffers from a lack of adaptability to net variations incurred by system reconfiguration, communication failure, and constraint integration. This article aims to enforce GMECs for a live PN model by using critical places and marking estimation. First, we define GMECs on some so-called critical places, such that the satisfaction of the GMECs can be determined by only monitoring the number of tokens in their corresponding critical places during runtime. Then, an efficient and effective control strategy is developed such that the controllers forbid all those transition firings that lead to the violated GMECs based on the estimated markings of critical places derived by observers from a resource perspective rather than exploring the markings of the entire system. Finally, we present procedures to deal with decision deadlocks, which are induced by one-sided decisions made by some controllers and may prevent all enabled transitions from firing. Global GMECs are always implemented through the local observation and control of processes without knowing an extra information.

In a multidomain application environment, it is of paramount importance for different organizations to collaborate with each other to facilitate secure interoperation. However, various types of conflicts related to access control constraints may arise as a result of integrating access control policies for individual domains, such as role inheritance violations (RIVs) and separation of duty violations (SoDVs). Current methods solve the conflicts in a centralized way by withdrawing or removing all crossdomain relationships resulting in the violations with the knowledge of all domains. However, these methods are inappropriate for large-scale systems due to their high computational complexity. In this article, we propose a distributed approach to avoid secure conflicts in a multidomain environment. We first model the role inheritance hierarchies of multiple domains as an interoperation graph. We then develop RIVs and SoDVs avoidance algorithms based on the interoperation graph and the communications among different domains. Each domain can execute the algorithms autonomously and in real time by evaluating whether its succeeding activated role can result in RIVs and SoDVs. We show that the new algorithms perform well in contrast to the existing algorithms.

Nowadays, workflow process changes frequently in a fast-changing business environment. When updating a workflow process via some structural changes, one of the most important tasks is to maintain its consistency under temporal constraints. Several approaches have been developed to cope with this issue. However, they are either inaccurate in locating changed parts or inefficient in analyzing temporal constraints owing to their excessive or erroneous estimation of affected portions for the updated workflow processes. Based on a sprouting graph (a graph that records the structure and time information of all paths in a workflow process in advance), this paper proposes a novel approach to analyzing temporal constraints for workflow processes with dynamic changes. First, changed parts are located via affected collaboration execution paths in the new model, i.e., the workflow process after changes. Second, instead of updating all the elements, only necessary (changed) nodes corresponding to the changed parts are updated in the sprouting graph of the original model, i.e., the workflow process before changes. Finally, based on the updated sprouting graph, only affected temporal constraints (the temporal constraints whose partial or all paths are contained in the affected collaboration execution paths) are checked. Compared with the existing works, our approach is applicable and efficient to check temporal constraints for large-scale and complex workflow processes thanks to its much lower time and space complexity.

Recently, the mediation-aided approach is attracting more attention in Web service composition, in the meanwhile, temporal constraints are regarded as an important aspect to ensure the correctness and QoS in service compositions. This combination leads to a new challenge in analyzing the timed compatibility of mediation-aided service composition. Unfortunately, existing model checking based approaches are lack of the ability of transform mediation-aided service composition to Time Automata (TA) models, and suffer from state space explosion for large-scale and complex compositions. In this paper, we present a new model checking approach to analyzing timed compatibility. Firstly, mediation-aided service composition is automatically decomposed into fragments. Secondly, each fragment is transformed into a TA. Finally, the temporal constraints are checked by the queries of observing TAs. Compared with existing approaches, our approach is able to check timed compatibility of mediation-aided service composition, and is more efficient than them.