Hybrid Process Model Research Articles

ACCOUNT MANAGEMENT COLLECTIONS OF PURCHASE ORDER CARD HOLDERS OR GUARANTORS

This research determines the account management collections of purchase order card holders or guarantors and the effectiveness of different collection strategies employed by retail stores. The objective is to integrate the 59.89 percent e-payment systems as a preferred payment method from the reflected 58.4 percent over-the-counter payment methods that resulted in a high risk of uncollected accounts or commonly known as doubtful accounts. Henceforth, automated invoicing of the statement of accounts is highly suggested to prevent delay, the conventional sending of a statement of accounts or invoicing presented 54.5 percent of purchase order card holders or guarantors did not receive their statement of accounts, during COVID-19. To recover from the economic loss this research showed a significant level of 0.03 on account management collections, indicating the need for effective collection strategies. This highlights the significance of the employee’s relationship to the collection process, indicating a positive correlation between groups and their strategies. The findings suggest that purchase order card holders or guarantors are willing to pay using e-wallets. The correlation presents the relationship between six variables: income (I), standardized collections (SD), payment behavior (PB), communication (C), timely invoicing (TI), and follow-up procedures (FP). The results indicate that all variables have a positive correlation with each other. The recommendation is to create a hybrid process model to manage collections strategically, incorporating e-wallets and automated statements of accounts to improve overall business performance. KEYWORDS: Accounts Management, Collections, Purchase Order Card Holders or Guarantors, E-Wallets, Automation, Hybrid Process Model, Business Administration

Physics-guided neural networks with engineering domain knowledge for hybrid process modeling

As neural networks are more frequently used to solve problems in science and engineering, the methods used to incorporate scientific knowledge into these networks are becoming increasingly complex. This work breaks down these complicated techniques into a set of basic strategies which can easily be applied to diverse situations. Several novel neural networks are built using the categories laid out in this work. These networks are tested on simulated data from a continuous stirred tank reactor (CSTR) model to evaluate the advantages provided by each network. The three points demonstrated in this work are: (1) architectural hybrid models can speed up convergence and reduce the amount of data necessary to train a model; (2) adding a physics-guided loss function can improve model generalization and make models more physically consistent; (3) using physics-guided initialization and transfer learning improves accuracy and speeds up convergence, but can harm generalizability if used incorrectly.

Data-Driven Fault Detection in Industrial Batch Processes Based on a Stochastic Hybrid Process Model

This paper presents a novel fault detection approach for industrial batch processes. The batch processes under consideration are characterized by the interaction between discrete system modes and non-stationary continuous dynamics. Therefore, a stochastic hybrid process model (SHPM) is introduced, where process variables are modeled as time-variant Gaussian distributions, which depend on hidden system modes. Transitions between the system modes are assumed to be either autonomous or to be triggered by observable events such as on/off signals. The model parameters are determined from training data using expectation-maximization techniques. A new fault detection algorithm is proposed, which assesses the likelihoods of sensor signals on the basis of the stochastic hybrid process model. Evaluation of the proposed fault detection system has been conducted for a penicillin production process, with the results showing a significant improvement over the existing baseline methods.Note to Practitioners—Automatic fault detection makes it possible to limit the effects of faults by taking countermeasures at an early stage. In this work, a data-driven fault detection method for industrial batch processes is proposed, in which the underlying process model is learned from training data. The proposed fault detection system can be used for various industrial batch processes without the need for complex and error-prone manual configuration. In contrast to many other data-driven approaches such as neural networks, only a few process cycles are required to create a robust process model. It should be noted that in data-driven fault detection methods, the training data should cover a large part of the process states that occur during error-free process cycles. The developed method is therefore particularly suitable for cyclical processes, which, however, can have alternative process paths and variability between the process cycles.

Hybrid process model and smart policy network of electric-vehicle resources for instantaneous power flow imbalances

The technology development of power electronics and battery empowers electric vehicles as practical approaches for the compensation of instantaneous power flow imbalance. To utilize the response capability of distributed vehicles and stabilize the grid frequency caused by instant power mismatch, a process model hybridizing the data-driven and conventional modes is designed for the electrical-grid-electric-vehicle system. Due to the real-time variations of power system, electric vehicles, and renewable energies, the hybrid process model directly interacts with the system feedback to remove the sophisticated model establishment, which effectively establishes the mapping relationship between system states and intelligently processing demands. Based on the hybrid process model, the smart policy network with an adversarial mechanism is then developed to enhance the model behavior. To fulfill the continuous action requirement and speed up the policy convergence, a proximal optimization strategy is further introduced to adjust hyperparameters through a stochastic ratio automatically. Bridging the huge volume of real-time dynamics and the action domain, the sufficient utilization of various operation states is achieved in the proposed process model and policy network, which notably detects the optimal operation point and reduces the power flow imbalance. The developed hybrid process model and smart policy network are validated through an electrical-grid-electric-vehicle system by comprehensive studies from four aspects of the process performance, generalization, robustness, and efficiency. Compared with the linear and naive hybrid process methods, 3.95 times superior integration performance and the improved evolution process of 64.5% improvement are both demonstrated as well as the robustness facing various circumstances.

A hybrid science‐guided machine learning approach for modeling chemical processes: A review

AbstractThis study presents a broad perspective of hybrid process modeling combining the scientific knowledge and data analytics in bioprocessing and chemical engineering with a science‐guided machine learning (SGML) approach. We divide the approach into two major categories: ML complements science, and science complements ML. We review the literature relating to the hybrid SGML approach, and propose a systematic classification of hybrid SGML models. For applying ML to improve science‐based models, we present expositions of direct serial and parallel hybrid modeling and their combinations, inverse modeling, reduced‐order modeling, quantifying uncertainty in the process and even discovering governing equations of the process model. For applying scientific principles to improve ML models, we discuss the science‐guided design, learning and refinement. For each subcategory, we identify its requirements, strengths, and limitations, together with their published and potential applications. We also present several examples to illustrate different hybrid SGML methodologies for modeling chemical processes.

Hybrid Process Modeling of an Industrial Process

AbstractThe current revival and intensified research on machine learning take place while an intensified focus of companies on digitalization developed. The willingness and acceptance to apply innovative methods in industry has grown. Gray box or hybrid process modeling represents the combination of rigorous (white box) process modeling with machine learning (black box) models. In recent applications, the presented method has been utilized for process models at Evonik. This article describes the workflow for developing hybrid process models and is meant to share the experiences made during the process.

THE CHALLENGING COMBINATION OF AGILITY AND CONVERGENCE IN HYBRID PRODUCT DEVELOPMENT PROCESSES: AN EMPIRICAL ANALYSIS OF STANFORD'S ME310 PROCESS MODEL

AbstractCombining agility and convergence in the development of physical products is a major challenge. Rooted in a design thinking approach, Stanford's ME310 process model attempts to resolve the conflicting priorities of these two design principles. To investigate how successful Stanford's hybrid process model is in doing so, we have used a qualitative case study approach. Our paper begins by outlining this process model's fundamental principles in terms of engineering design methodology. Subsequently, we present the results of our empirical analysis, which tracks the coevolution of problem and solution space by meticulously examining all prototype paths in ten of Stanford's ME310 student projects. We have discovered that convergence during solution finding does not correspond to the process model's theoretical specifications. Even in the phase of the final prototype, both the technical concept and the underlying problem formulation changed frequently. Further research should focus on combining the prototype-based ME310 approach with methods from systems engineering which allow for a more comprehensive theoretical exploration of the solution space. This could lead to improved convergence during solution development.

Cost evaluation and optimisation of hybrid multi effect distillation and reverse osmosis system for seawater desalination

In this research, the effect of operating parameters on the fresh water production cost of hybrid Multi Effect Distillation (MED) and Reverse Osmosis (RO) system is investigated. To achieve this, an earlier comprehensive model developed by the authors for MED + RO system is combined with two full-scale cost models of MED and RO processes collected from the literature. Using the economic model, the variation of the overall fresh water cost with respect to some operating conditions, namely steam temperature and steam flow rate for the MED process and inlet pressure and flow rate for the RO process, is accurately investigated. Then, the hybrid process model is incorporated into a single-objective non-linear optimisation framework to minimise the fresh water cost by finding the optimal values of the above operating conditions. The optimisation results confirm the economic feasibility of the proposed hybrid seawater desalination plant.

Deterministic global process optimization: Accurate (single-species) properties via artificial neural networks

Global deterministic process optimization problems have recently been solved efficiently in a reduced-space by automatic propagation of McCormick relaxations (Bongartz and Mitsos, J. Global Optim, 2017). However, the previous optimizations have been limited to simplified thermodynamic property models. Herein, we propose a method that learns accurate thermodynamic properties via artificial neural networks (ANNs) and integrates those in deterministic global process optimization. The resulting hybrid process model is solved using the recently developed method for deterministic global optimization problems with ANNs embedded (Schweidtmann and Mitsos, J. Optim. Theory Appl., 2018). The optimal operation of a validated steady state model of an organic Rankine cycle is solved as a case study. It is especially challenging as the thermodynamic properties are given by the implicit Helmholtz equation of state. The results show that modeling of thermodynamic properties via ANNs performs favorable in deterministic optimization. This method can rapidly be extended to include properties from existing thermodynamic libraries, based on models or data.

A hybrid description and evaluation of oxymethylene dimethyl ethers synthesis based on the endothermic dehydrogenation of methanol

Concerning oxymethylene dimethyl ethers (OME), this work utilises a hybrid process model based on methanol (MeOH) and its partially selective conversion to anhydrous formaldehyde (FA), which in turn is used as the feed for OME synthesis.

Hybrid simulation of laser deep penetration welding

AbstractIn laser deep penetration welding, the knowledge on the temperature history of the material is of great interest for the assessment of the quality properties of the weld. For this purpose a hybrid process model that enables the fast calculation of temperature distributions as a function of process parameters is applied. The interaction between laser and material is taken into account by a reduced keyhole model, which exploits a hierarchy in the spatial dimensions occurring at high feed rates. The resulting shape of a stationary keyhole is introduced as a Dirichlet boundary into a thermal finite element simulation in which it is moved through the workpiece according to the process control of the laser beam. The boundary is mathematically described by a level set function and immersed in a fixed computational mesh. The Dirichlet boundary condition is imposed using an embedded boundary method. The calculated temperature distributions are evaluated by means of bead on plate welds conducted in 0.9 mm thick sheets of 1.4301 (AISI 304) stainless steel.

Growth decline assessment in Pinus sylvestris L. and Pinus nigra Arnold. forest by using 3-PG model.

Aim of the study: We assessed the ability of the 3-PG process-based model to accurately predict growth of Pinus sylvestris and P. nigra plantations across a range of sites, showing declining growth trends, in southern Spain.Area of study: The study area is located in “Sierra de Los Filabres” (Almería).Material and methods: The model was modified in fifteen parameters to predict diameter (DBH, cm), basal area increment (BAI, cm2 yr-1) and leaf area index (LAI, m2 m-2) in healthy trees and trees showing declining growth. We assumed that a set of specific physiological parameters (stem partitioning ratio-pFS20, maximum litterfall rate-γFx, maximum canopy conductance-gCx, specific leaf area for mature aged stands-σ1, age at which specific leaf area = ½ (σ0 + σ1), age at full canopy cover-tc, and canopy boundary layer conductance-gB) included in 3-PG would be suitable for predicting growth decline related to climate conditions. The calibrated model was evaluated using dendrochronological and LAI data obtained from plots.Main results: Observed and simulated DBH showed a high correlation (R2 > 0.99) between modelled and measured values for both species. In contrast, modelled and observed BAI showed lower correlation (R2 < 0.68). Sensitivity analysis on 3-PG outputs showed that the foliage parameters - maximum litterfall rate, maximum canopy conductance, specific leaf area for mature aged stands, age at which specific leaf area, and age at full canopy cover - were important for DBH and BAI predictions under drought stress.Research highlights: Our overall results indicated that the 3-PG model could predict growth response of pine plantations to climatic stress with desirable accuracy in southern Spain by using readily available soil and climatic data with physiological parameters derived from experiments.Keywords: Hybrid process model; forest management models; growth prediction; Pinus spp, Parameterization; forest decline.

A hybrid process model for EDM based on finite-element method and Gaussian process regression

This paper proposed a hybrid intelligent process model, based on finite-element method (FEM) and Gaussian process regression (GPR), for electrical discharge machining (EDM) process. A model of single-spark EDM process has been constructed based on FEM method, considering the latent heat, variable heat distribution coefficient of cathode (f c ), and plasma flushing efficiency (PFE), to predict material removal rate (MRR) and surface roughness (Ra). This model was validated using reported analytical and experimental results. Then, a GPR model was proposed to establish relationship between input process parameters (pulse current, pulse duration, and discharge voltage) and the process responses (MRR and Ra) for EDM process. The GPR model was trained, tested, and tuned using the data generated from the numerical simulations. Through the GPR model, it was found that responses of EDM process can be accurately predicted for the chosen process conditions. Therefore, for the selection of optimum parameters, the hybrid intelligent model proposed in this paper can be used in EDM process.

Predicting Under Nutrition Status of Under-Five Children Using Data Mining Techniques: The Case of 2011 Ethiopian Demographic and Health Survey

Background: Under nutrition is one of the leading causes of morbidity and mortality in children under the age of five in most developing countries including Ethiopia. The main objective of this study was to design a model that predicts the nutritional status of under-five children using data mining techniques. Methods: This study followed hybrid methodology of Knowledge Discovery Process to achieve the goal of building predictive model using data mining techniques and used secondary data from 2011 Ethiopia Demographic and Health Survey (EDHS) dataset. Hybrid process model was selected since it combines best features of Cross-Industry Standard Process for Data Mining and Knowledge Discovery in Database methodology to identify and describe several explicit feedback loops which are helpful in attaining the research objectives. WEKA 3.6.8 data mining tools and techniques such as J48 decision tree, Naive Bayes and PART rule induction classifiers were utilized as means to address the research problem. Result: In this particular study, the predictive model developed using PART pruned rule induction found to be best performing having 92.6% of accurate results and 97.8% WROC area. Promising result has been achieved from the rules regarding nutritional status prediction. Conclusion: The results from this study were encouraging and confirmed that applying data mining techniques could indeed support a predictive model building task that predicts nutritional status of under-five children in Ethiopia. In the future, integrating large demographic and health survey dataset and clinical dataset, employing other classification algorithms, tools and techniques could yield better results.

Improving Environmental Scanning Systems Using Bayesian Networks

AbstractAs companies’ environment is becoming increasingly volatile, scanning systems gain in importance. We propose a hybrid process model for such systems’ information gathering and interpretation tasks that combines quantitative information derived from regression analyses and qualitative knowledge from expert interviews. For the latter, we apply Bayesian networks. We derive the need for such a hybrid process model from a literature review. We lay out our model to find a suitable set of business environment indicators to forecast a company’s key financials. Deriving lessons learned from a prototype in the industrial sector, we evaluate the utility of our model following the design science research paradigm. We find our model to especially convince in completeness, transparency and transportability when compared with “pure” mathematical models.

Modeling and optimization of cross-flow ultrafiltration using hybrid neural network-genetic algorithm approach

Precise modeling flux decline under various operating parameters in cross-flow ultrafiltration (UF) of oily wastewaters and afterward, employing an appropriate optimization algorithm in order to optimize operating parameters involved in the process model result in attaining desired permeate flux, is of fundamental great interest from an economical and technical point of view. Accordingly, this current research proposed a hybrid process modeling and optimization based on computational intelligence paradigms where the combination of artificial neural network (ANN) and genetic algorithm (GA) meets the challenge of specified-objective based on two steps: first the development of bio-inspired approach based on ANN, trained, validated and tested successfully with experimental data collected during the polyacrylonitrile (PAN) UF process to treat the oily wastewater of Tehran refinery in a laboratory scale in which the model received feed temperature (T), feed pH, trans-membrane pressure (TMP), cross-flow velocity (CFV), and filtration time as inputs; and gave permeate flux as an output. Subsequently, the 5-dimensional input space of the ANN model portraying process input variables was optimized by applying GA, with a view to realizing maximum or minimum process output variable. The results obtained validate the estimates of the ANN–GA technique with a good accuracy. Finally, the relative importance of the controllable operation factors on flux decline is determined by applying the various correlation statistic techniques. According to the result of the sensitivity analysis based on the correlation coefficient, the filtration time was the most significant one, followed by T, CFV, feed pH and TMP.

The rationality of emotions: A hybrid process model of decision-making under uncertainty

Decision-making continues to be a vital area of research on organizations, particularly in Asia. After years of research under the cognitive approach, there has been an upsurge of scholarly attention connecting decision-making to emotion. Nonetheless, competing accounts of the role played by emotion in decision-making have created a conundrum as to whether emotion is rational or irrational and whether it should even be considered along with other cognitive aspects of the decision-making process. In response, we derive a hybrid process decision-making model to integrate the effects from cognition and emotions varying with the levels of uncertainty. In proposing this model, we seek to differentiate the rationality of emotions from the instrumental, functional, and expressive mechanisms in decision-making. In particular, we proffer that affective construal legitimates the rational account of emotions under high uncertain situation. Finally, we draw some key implications and recommendations for organizational research in the Asian region.

A Hybrid Process Modeling Method and its Application in Equipment Support System

In this paper, a hybrid process modeling method for equipment support system was presented. Combining with the advantages of ARENA and UML, the models of equipment support process were constructed by the means of UML, which include function models, dynamic models and object models. The frame work of UML-ARENA simulation software was designed, and the mapping relations between UML object classes and ARENA models were founded. The equipment support process was simulated by ARENA. At last, an example was given, which proves that this hybrid modeling method is feasible and significant for its can optimize the support process, shorten the support time, and improve the usage ratio of resources.

Impact of alkalinity sources on the life-cycle energy efficiency of mineral carbonation technologies

This study builds a holistic, transparent life cycle assessment model of a variety of aqueous mineral carbonation processes using a hybrid process model and economic input–output life cycle assessment approach (hybrid EIO-LCA). The model allows for the evaluation of the tradeoffs between different reaction enhancement processes while considering the larger lifecycle impacts on energy use and material consumption. A preliminary systematic investigation of the tradeoffs inherent in mineral carbonation processes is conducted to provide guidance for the optimization of the life-cycle energy efficiency of various proposed mineral carbonation processes. The life-cycle assessment of aqueous mineral carbonation suggests that a variety of alkalinity sources and process configurations are capable of net CO2 reductions. The total CO2 storage potential for the alkalinity sources considered in the U.S. ranges from 1.8% to 23.7% of U.S. CO2 emissions, depending on the assumed availability of natural alkalinity sources and efficiency of the mineral carbonation processes.

A Process Model of Collaborative Management Research: The Study of Collective Creativity in the Luxury Industry

This article explores collaborative management research (CMR). As one of the streams within the action research family, CMR is one approach that has been identified as a potent method for advancing scientific knowledge and bringing about change in organizations. The article proposes a hybrid model of the CMR research process in organizations. Following a brief introduction of collaborative management research, we advance an emerging, inductively derived, hybrid process model of CMR. Three critical clusters for achieving the intended outcomes of collaborative management research include: (1) contextual factors, (2) quality of the collaboration, and (3) the development of the collaborative research process itself. Within each cluster, concepts, variables, and processes were identified and linked together to form a hybrid model of CMR process. An illustration of a collaborative management research project that focused on the study of collective creativity with an Italian fashion and design company is followed by a reflective analysis. The discussion provides directions for future research and implications for practice.