Information Of Orders Research Articles

A Model-Agnostic Graph Neural Network for Integrating Local and Global Information

Graph Neural Networks (GNNs) have achieved promising performance in a variety of graph-focused tasks. Despite their success, however, existing GNNs suffer from two significant limitations: a lack of interpretability in their results due to their black-box nature, and an inability to learn representations of varying orders. To tackle these issues, we propose a novel Model-agnostic Graph Neural Network (MaGNet) framework, which is able to effectively integrate information of various orders, extract knowledge from high-order neighbors, and provide meaningful and interpretable results by identifying influential compact graph structures. In particular, MaGNet consists of two components: an estimation model for the latent representation of complex relationships under graph topology, and an interpretation model that identifies influential nodes, edges, and node features. Theoretically, we establish the generalization error bound for MaGNet via empirical Rademacher complexity, and demonstrate its power to represent layer-wise neighborhood mixing. We conduct comprehensive numerical studies using simulated data to demonstrate the superior performance of MaGNet in comparison to several state-of-the-art alternatives. Furthermore, we apply MaGNet to a real-world case study aimed at extracting task-critical information from brain activity data, thereby highlighting its effectiveness in advancing scientific research.

Multi-view unsupervised complementary feature selection with multi-order similarity learning

Graph construction, an open and challenging problem, is of great significance for multi-view unsupervised feature selection. So far, many graph construction methods, such as distance-based (the local structure) and self-reconstruction-based (the global structure), have been devised to serve the feature selection task. Although these methods have achieved some improvements, they fail to utilize high-order neighbor information, let alone exploit the neighbor information of different orders, to improve the feature selection task. In this paper, we propose a new insight to construct graphs that can accommodate multi-order neighbor information for selecting the relevant features. Besides, we observe that existing methods adopts the general information fusion strategy in multi-view learning, e.g. fusing graphs, without taking into account the unique characteristics of the feature selection task. Therefore, the proposed method seeks to project multi-view data onto a shared latent representation, which explores the complementarity tailored to the feature selection task at the feature level. A simple yet effective algorithm is designed to solve the optimization problem of the objective function. Extensive clustering experiments demonstrate the superiority of our method over state-of-the-art ones.

Graph transformer-based self-adaptive malicious relation filtering for fraudulent comments detection in social network

In recent years, Graph Neural Networks (GNNs) have proven to be effective in detecting fraud within social networks by gathering information from neighboring nodes to predict fraudulent actions. However, the continuous evolution and camouflaging tactics of fraudsters, such as establishing malicious connections with legitimate users and mimicking their behaviors, can often elude GNN-based detection methods. To tackle this issue, we propose a multi-order moments graph transformer, named MMGT, to effectively learn node representations through an attention mechanism that captures information of different orders of moments. Building upon this foundation, we further introduce a self-adaptive malicious relation filtering model for fraud detection, denoted as SFGT. Initially, we form a multi-relational graph that encapsulates complex relations within a given social network and derive node representations by accumulating neighbor node and edge data based on an enhanced graph transformer. Subsequently, a threshold-based malicious relation filter mechanism is proposed to eliminate malicious links by assessing the distance between nodes. Furthermore, an adaptive threshold learning policy is developed to bolster the model’s performance and its ability to generalize. Finally, extensive experiments are carried out on two public datasets, Amazon and YelpChi, which underscores the effectiveness of our proposed model. The experimental results indicate that our model achieves state-of-the-art performance.

Multi-order graph attention network for water solubility prediction and interpretation

The water solubility of molecules is one of the most important properties in various chemical and medical research fields. Recently, machine learning-based methods for predicting molecular properties, including water solubility, have been extensively studied due to the advantage of effectively reducing computational costs. Although machine learning-based methods have made significant advances in predictive performance, the existing methods were still lacking in interpreting the predicted results. Therefore, we propose a novel multi-order graph attention network (MoGAT) for water solubility prediction to improve the predictive performance and interpret the predicted results. We extracted graph embeddings in every node embedding layer to consider the information of diverse neighboring orders and merged them by attention mechanism to generate a final graph embedding. MoGAT can provide the atomic-specific importance scores of a molecule that indicate which atoms significantly influence the prediction so that it can interpret the predicted results chemically. It also improves prediction performance because the graph representations of all neighboring orders, which contain diverse range of information, are employed for the final prediction. Through extensive experiments, we demonstrated that MoGAT showed better performance than the state-of-the-art methods, and the predicted results were consistent with well-known chemical knowledge.

Remote control and monitoring of a hydraulic machine

An electronic system based on an open source programmable board was designed and tested, for remote control and monitoring of a fluid power machine. To satisfy the requirements for remote two-way communication between the machine and the user interface, a connection must be established via the Internet. The Arduino Uno Wi-Fi Rev 2 board is designed to connect wirelessly to a local network. To transmit the data between the board and the graphical interface, on a device that is on an external network, a MQTT broker was generated on an AWS server. A program receives the information from the sensors, sends it to the user interface, makes logical decisions, processes the information of control orders, and establishes the protocols to govern the machine. The electronic system was tested with a mobile hydraulic crane. Four variables were selected to control and monitor the crane performance: the load position, the current demanded by the electric motor, the flow supplied by the hydraulic pump and the presence of mechanical pins to lock or release the hydraulic linear actuator. Two tests were conducted to control the machine remotely, one within the same location and the other from Berlin, Germany. It was possible to use the open source board for connecting with the user interface generated with Node-RED via MQTT protocol and control and monitor the machine performance.

Key Frame Extraction of Surveillance Video Based on Fractional Fourier Transform

With the vigorous development of national infrastructure construction and public information construction, video surveillance systems have gradually penetrated various fields. The current key frame extraction technology has inadequate target details and inaccurate judgment of local actions. Addressing this problem, a key frame extraction method based on fractional Fourier transform is proposed. This method obtained the phase spectra information of different orders by performing fractional Fourier transform on the surveillance video frames. Next, the method designed an adaptive algorithm based on the golden section point to select the transformation order. Then, the phase spectrum information of two adjacent frames was used to characterize the changes in the global and local motion states of the target. The final step was to extract key frames based on this. Experimental results show that, compared with the previous methods, the key frames extracted by the method proposed in this paper can correctly capture the changes in the global and local motion states of the target.

WebGIS-Based Collaborative Construction Quality Control of RCC Gravity Dam Using Sensing Devices

In recent years, information technology has been widely used in the construction quality control of hydropower stations, and a series of construction management systems based on the internet of things (IOT) have been produced. However, the existing systems are usually aimed at specific construction quality control factors, leading to the independence of all kinds of monitoring information, which also makes it impossible to directly evaluate the causes of quality problems after they occur. Manual multifactor analysis and evaluation are still needed. Based on a roller compacted concrete (RCC) gravity dam as an example, the construction quality control method is achieved by using WebGIS technology, image recognition technology, voxelization technology, and data mining analysis, which can quickly evaluate the causes of quality problems and integrate all kinds of monitoring information of different orders of magnitude into a presentation scene. This method makes the project quality management process more effective and cooperative. This study contributes to the existing body of knowledge of the quality control method based on informatization in construction.

An Adversarial Deep Hybrid Model for Text-Aware Recommendation with Convolutional Neural Networks

The standard matrix factorization methods for recommender systems suffer from data sparsity and cold-start problems. Thus, in real-world scenarios where items are commonly associated with textual data such as reviews, it becomes necessary to build a hybrid recommendation model that can fully utilize the text features. However, existing methods in this area either cannot extract good features from the texts due to their order–insensitive document modeling approaches or fail to learn the hybrid model in an effective way due to their complexity of inferring the latent vectors. To this end, we propose a deep hybrid recommendation model which seamlessly integrates matrix factorization with a Convolutional Neural Network (CNN), a powerful text feature extraction tool with the capability of detecting the information of word orders. Unlike previous works which use content features as prior knowledge to regularize the latent vectors, we combine CNN into MF in an additive manner to allow training CNN with direct learning signals. Furthermore, we propose an adversarial training framework to learn the hybrid recommendation model, where a generator model is built to learn the distribution over the pairwise ranking pairs while training a discriminator to distinguish generated (fake) and real item pairs. We conduct extensive experiments on three real-world datasets to demonstrate the effectiveness of our proposed model against state-of-the-art methods in various recommendation settings.

Replenishment policy for a purchase-to-order seller: a tradeoff between ordering cost and delay cost

We consider replenishment decisions on when and how many goods to purchase for a seller under a purchase-to-order mode where there is no inventory and the seller purchases goods to fulfil orders already placed. For each purchase, there is a constant ordering cost. For each order, delay cost will be incurred if it is not fulfilled timely. Generally, the more frequent the replenishment, the higher the ordering cost but the lower the delay cost. Consequently, there is a tradeoff between the ordering cost and the delay cost for the seller to make replenishment decisions minimising the total cost. In this paper, we study three cases of the problem and investigate both offline versions and online versions according to the seller's knowledge about information of future orders. For offline versions with perfect information, we either develop an optimal policy, or prove it is NP-hard and develop an approximation policy. For online versions without any information about future orders, from the perspective of competitive analysis we prove the lower bound of competitive ratio for any possible online policy and present a 10-competitive online policy for the general case and a 2-competitive online policy for each of two special cases.

Improving Order Execution Cost Estimation by Semisupervised Learning

Order execution cost analysis is one of the most important problems in financial investments. Many previous research works model the problem as a cost classification or regression task. However, due to insufficient real orders, performances of those models are not satisfying. Moreover, unlimited simulated orders generated by market simulators are not exploited by the analysis approach. In this paper, we propose an order execution cost estimation approach by using limited real orders and unlimited simulated orders. The approach 1) employs exploratory data analysis to explore the patterns and relationships included in the raw data, and selects the appropriate features for model training, 2) trains supervised models on labeled orders as baselines to estimate order execution cost, 3) trains three Semisupervised Learning (SSL) models on both labeled and simulated orders to improve the estimation performances, where a. Semisupervised Support Vector Machine (S3VM) makes a low-density separation on labeled and unlabeled orders, b. Tri-Training performs bootstrap sampling on the labeled orders to obtain three labeled training sets to make disagreement for labeling unlabeled orders, and c. Label Propagation (LP) model propagates the order execution cost labels of the labeled orders to the unlabeled ones on a graph and adjusts the labels based on local and global consistency. Experiments are conducted on real and simulated order datasets. Results of the experiments show that the SSL models perform better than the baselines, where S3VM optimized by Adam, Random Forest (RF) based Tri-Training and Radial Basic Function (RBF) based LP can make use of the information of unlabeled orders to tremendously improve classification performances in F1 score.

Work orders management based on XML file in printing

The Extensible Markup Language (XML) technology is increasingly used in various field, if it’s used to express the information of work orders will improve efficiency for management and production. According to the features, we introduce the technology of management for work orders and get a XML file through the Document Object Model (DOM) technology in the paper. When we need the information to conduct production, parsing the XML file and save the information in database, this is beneficial to the preserve and modify for information.

Ordered visibility graph average aggregation operator: An application in produced water management.

Complex networks are widely used in modeling complex system. How to aggregate data in complex systems is still an open issue. In this paper, an ordered visibility graph average aggregation operator is proposed which is inspired by the complex network theory and Newton's law of universal gravitation. First of all, the argument values are ordered in descending order. Then a new support function is proposed to measure the relationship among values in a visibility graph. After that, a weighted network is constructed to determine the weight of each value. Compared with the other operators, the new operator fully takes into account not only the information of orders but also the correlation degree between the values. Finally, an application of produced water management is illustrated to show the efficiency of the proposed method. The new method provides a universal way to aggregate data in complex systems.

Construction Material Safety-Stock Determination Under Nonstationary Stochastic Demand and Random Supply Yield

Nonstationary stochastic demand and random supply yield have led to a great challenge in construction material management. Appropriate safety stocks can cope with uncertainty in both demand and supply yield, and improve the construction cost and productivity significantly. But the safety stock is usually determined by planner's experience or rough evaluation in management practice, which results in overstock or understock situations frequently. For changing this situation, this paper focuses on analysis of the stochastic inventory control problem of construction material, and develops a general safety-stock determination approach under nonstationary stochastic demand and random supply yield. Based on the linear-order release rule and modified base-stock definition, the inventory balance equation is developed. Then, a coverage random variable, which incorporates the complete information of outstanding orders, is defined to represent the outflows and inflows to stock over risk periods. Meanwhile, the fixed-point iteration method is adopted to calculate the time-varying safety-stock level under predefined service-level (nonstockout probability) constraint. Finally, a numerical analysis is conducted to investigate the relative performance of the proposed safety-stock approach and days of supply rule. The result demonstrates a considerable service improvement and inventory decrease of the safety-stock approach.

Ordered visibility graph weighted averaging aggregation operator: A methodology based on network analysis

The ordered weighted averaging (OWA) operators are widely used in many applications. Determining the OWA operator weights is still an open issue. In this paper, an ordered visibility graph weighted averaging (OVGWA) aggregation operator is proposed based on complex network methodology. After the argument values are ordered in descending order, a network is constructed using ordered visible graph (OVG) method. The arriving probabilities of random walker in the network are utilized as the weights of the corresponding data. Compared with the classical OWA, the proposed OVGWA not only takes the information of orders but also the argument values themselves into consideration to determine the weights. An application of produced water management is used to illustrate the application of the proposed OVGWA aggregation operator. In addition, the aggregated results are compared with those obtained by maximal entropy method. The results show that the proposed method is effective. To study the impacts of the influencing factors, a sensitivity analysis has been carried out. It is utility to understand the role of different indicators on aggregated value.

Portfolio credit risk with predetermined default orders

Portfolio credit risk models can be distinguished by the use of a top-down approach or a bottom-up one. The main difference between these two approaches is the information of default identities. In this paper, we propose a conditional top-down approach which models the default times with a predetermined default order of identities. Thus conditioned on the default order, the default times of a bottom-up model can be constructed simply using a top-down approach. We use the tool of assumptions to separate the information of default orders from the ordered default times. The predetermined assumption (a special assumption) introduced here allows that the construction of the loss process relates to a probability on permutations. We can derive the probabilities on default orders from the known bottom-up models satisfying the predetermined assumption (e.g. Jarrow-Yu’s contagion model), and obtain new choices of probability on default orders based on some simple and interesting indices of permutations such as the inverse index. Furthermore, under the predetermined assumption, some generic pricing problems of the bottom-up models can be simplified to the special case of the conditional Markov loss model. We then apply these results to Jarrow-Yu’s contagion model and give an efficient approach to the pricing problem of CDO tranches, where new expansions of the loss distributions are derived by the random matrix exponential.

Simulation and experimental design methods for job shop scheduling with material handling: a survey

Job shop scheduling (JSS) problems have been studied for over six decades. Many of them are proved to be non-deterministic polynomial-time (NP) hard, which means that they are intractable and the computation time increases exponentially with the problem size goes up. Some assumptions have been made in the previous studies about JSS problems in order to simplify the model and solve it. One of those assumption is that the transferring times of jobs between different machines are negligible. However, it is highly not practical not to model the material handling activities in a typical shop floor scheduling problems in reality, especially when the movements of jobs on shop floor are completely relying on the material handling equipment and the transferring times are comparable to the production times. Omitting the transferring times will make the result of scheduling impossible to be implemented. Therefore, many recent studies have been done about Job shop scheduling with material handling (JSPMH/JSSMH). The problem has been defined into two different categories: offline and online scheduling problems. Offline scheduling means solving JSSMH problem as static scheduling problem that is solved before shop floor starts its production with known or predicted information of orders, while online scheduling is dynamic scheduling problem that is solved to generate schedules in real time as orders arrive on the shop floor. In both methods, the simulation modeling tool has been widely used with its ability of efficiently searching for optimal solution as well as evaluating the results. In articles that study the JSSMH problems as the dynamic scheduling, simulation method has been popularly used to factor in various scenarios of production. Our review will comprehensively summarize how JSSMH problems are solved in dynamic and static problem settings, as well as how simulation models play a useful role in solving this type of problems.

Dynamic Vehicle Routing Problem with Multiple Depots

The Vehicle Routing Problems (VRPs) has been extensively studied and applied in many fields. Variations of VRPs have been proposed and appeared in research for many decades. Dynamic Vehicle Routing Problem with Multiple Depots (D-MDVRP) extends the variation of VRPs to dynamism of customers by knowing the information of customers (both locations and due dates) at diverse times. An application of this problem can be found in food delivery services which have many service stores. The customer delivery orders are fulfilled by a group of scattered service stores which can be analogous to depots in D-MDVRP. In this example the information of all customer orders are not known at the same time depending on arrivals of customers. Thus the objective of this operation is to determine vehicle routing from service stores as well as dispatching time. This paper aims to develop a heuristic approach for D-MDVRP. The proposed heuristic method comprises of two phases: route construction and vehicle dispatch. Routes are constructed by applying the Nearest Neighbor Procedure (NNP) to cluster customers and select a proper depot, Sweeping and Reordering Procedures (SRP) to generate initial feasible routes, and Insertion Procedure (IP) to improve routing. Then the determination of dispatch is followed in the next phase. In order to deal with the dynamism, the dispatch time of each vehicle is determined by maximizing the waiting time to provide the opportunity to add more arriving customers in the future. An iterative process between two phases is adopted when a new customer enters the problem, and the vehicles are dispatched when the time becomes critical. From the computational study, the heuristic method performs well on small sized test problems in a shorter CPU time compared to the optimal solutions from CPLEX, and provides an overall average of 8.36 % Gap. For large size test problems, the heuristic method is compared with static problems, and provides an overall average of 3.48 % Gap.

Will Rest for the Weary Wear Everybody Out?

Will Rest for the Weary Wear Everybody Out?

Scheduling optimisation of co-operator selection and task allocation in mass customisation supply chain based on collaborative benefits and risks

How to deal with the contradiction between mass production effects and customised demand is the key problem of mass customisation (MC). With the development of supply chain management, we can probe a new way to solve this problem through the excellent character of supply chain scheduling (SCS). However, SCS in MC has special characters, mainly reflected in the random information of customer orders and the outstanding collaborative benefits and risks which can cause many complicated contradictions in operation. Based on earlier work about solving these contradictions, this paper introduces the collaborative benefit and risk analysis into scheduling and emphasises it not only for short-term but also for long-term collaboration. A dynamic and multi-objective optimisation model and an algorithm are set up for the scheduling optimisation of co-operator selection and task allocation. The model and algorithm not only reflect the operating requirements for the scheduling selection process in MC, but also reflect the thought of solving the key problem in MC and give consideration to the collaborative benefits and risks. The application feasibility of the model and algorithm are ultimately validated through an application case study.

Efficient Data Accessing through Heterogeneous ERP Solution

Application of Information Technology like ERP, Ecommerce, Data warehousing and Data mining are getting popularity in certain areas of industrial operations like Sales, production & Production planning, EIS(Executive Information System),System Admin, Statutory requirements, Corporate portal, Customer care, maintenance, quality control, inventory management, finance, marketing and HRIS(HR & Payroll). If we discuss to sustain competitive advantage in the dynamic market, an analysis at the micro level is essential e.g. Market and difference within markets, well-articulated understanding of market needs & implications of these for all functions i.e.. On time Delivery (OTD) of raw material market projections, order scheduling, production planning & control and after sales analysis apart from that big challenge is maintain the structured data like Multimedia and synchronization of access & retrieval of data according to final output. ERP is the only solution to face all the changes emerged due to changed business environment and also for the problems due to preventing traditional system of working in terms of data accessibility for the heterogeneous platform through the ERP solution. In Industry accurate planning and scheduling of orders, better date predictions, quick response to query and online detailed information of orders is must for doing business. As a result in the past few years at industry have invested in integrated systems, along with powerful hardware and software. The change from customized modular software to ERP is the current trend in Industry. Different type of industry are now going in for ERP/MRP (Material Resource Planning) and also web based software to enable them to remain competitive in this highly technology and competition driven economy. Effective application of Information technology in Industry has now become an integral part of their operations and a dire necessity for facing global challenges. Major benefits of ERP based systems are prompt and paperless communication, Better understanding of processes and business practices during implementation, an automatic discipline is imbibed in the users as lack of or delay in data entry by one department holds up the work in the next department. ERP helps in work center capacity analysis, work planning, advance planning of materials, stock management, warehouse management and reduced process cycle times resulting from using online purchase requisitions. By adopting ERP in industry system real time approach is possible to overcome the problem as mentioned above and also improve information system in terms of data access & retrieval.