Development of predictive models for effective process parameter selection for single and overlapping laser clad bead geometry

2018 ◽  
Vol 24 (1) ◽  
pp. 214-228 ◽  
Author(s):  
Kush Aggarwal ◽  
R.J. Urbanic ◽  
Syed Mohammad Saqib
Keyword(s):  
Data Collection ◽  
Predictive Model ◽  
Laser Cladding ◽  
Process Planning ◽  
Process Parameters ◽  
Predictive Models ◽  
Goodness Of Fit ◽  
Bead Geometry ◽  
Content Type ◽  
Non Linear

Purpose The purpose of this work is to explore predictive model approaches for selecting laser cladding process settings for a desired bead geometry/overlap strategy. Complementing the modelling challenges is the development of a framework and methodologies to minimize data collection while maximizing the goodness of fit for the predictive models. This is essential for developing a foundation for metallic additive manufacturing process planning solutions. Design/methodology/approach Using the coaxial powder flow laser cladding method, 420 steel cladding powder is deposited on low carbon structural steel plates. A design of experiments (DOE) approach is taken using the response surface methodology (RSM) to establish the experimental configuration. The five process parameters such as laser power, travel speed, etc. are varied to explore their impact on the bead geometry. A total of three replicate experiments are performed and the collected data are assessed using a variety of methods to determine the process trends and the best modelling approaches. Findings There exist unpredictable, non-linear relationships between the process parameters and the bead geometry. The best fit for a predictive model is achieved with the artificial neural network (ANN) approach. Using the RSM, the experimental set is reduced by an order of magnitude; however, a model with R2 = 0.96 is generated with ANN. The predictive model goodness of fit for a single bead is similar to that for the overlapping bead geometry using ANN. Originality/value Developing a bead shape to process parameters model is challenging due to the non-linear coupling between the process parameters and the bead geometry and the number of parameters to be considered. The experimental design and modelling approaches presented in this work illustrate how designed experiments can minimize the data collection and produce a robust predictive model. The output of this work will provide a solid foundation for process planning operations.

Using Predictive Modeling and Classification Methods for Single and Overlapping Bead Laser Cladding to Understand Bead Geometry to Process Parameter Relationships

2016 ◽  
Vol 138 (5) ◽  
Author(s):  
R. J. Urbanic ◽  
S. M. Saqib ◽  
K. Aggarwal
Keyword(s):  
Laser Cladding ◽  
Predictive Modeling ◽  
Predictive Models ◽  
Goodness Of Fit ◽  
Process Parameter ◽  
Bead Geometry ◽  
Process Characteristics ◽  
Single Bead ◽  
Reduced Gradient ◽  
Generalized Reduced Gradient

Developing a bead shape to process parameter model is challenging due to the multiparameter, nonlinear, and dynamic nature of the laser cladding (LC) environment. This introduces unique predictive modeling challenges for both single bead and overlapping bead configurations. It is essential to develop predictive models for both as the boundary conditions for overlapping beads are different from a single bead configuration. A single bead model provides insight with respect to the process characteristics. An overlapping model is relevant for process planning and travel path generation for surface cladding operations. Complementing the modeling challenges is the development of a framework and methodologies to minimize experimental data collection while maximizing the goodness of fit for the predictive models for additional experimentation and modeling. To facilitate this, it is important to understand the key process parameters, the predictive model methodologies, and data structures. Two modeling methods are employed to develop predictive models: analysis of variance (ANOVA), and a generalized reduced gradient (GRG) approach. To assist with process parameter solutions and to provide an initial value for nonlinear model seeding, data clustering is performed to identify characteristic bead shape families. This research illustrates good predictive models can be generated using multiple approaches.

Developing a predictive model of construction industry-university research collaboration

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Monty Sutrisna ◽  
Dewi Tjia ◽  
Peng Wu
Keyword(s):  
Data Collection ◽  
Predictive Model ◽  
Construction Industry ◽  
Research Collaboration ◽  
Ann Model ◽  
Data Set ◽  
Content Type ◽  
Construction Companies ◽  
Likelihood Model ◽  
Artificial Neural Network Ann

Purpose This paper aims to identify and examine the factors that influence construction industry-university (IU) collaboration and develop the likelihood model of a potential industry partner within the construction industry to collaborate with universities. Design/methodology/approach Mix method data collection including questionnaire survey and focus groups were used for data collection. The collected data were analysed using descriptive and inferential statistical methods to identify and examine factors. These findings were then used to develop the likelihood predictive model of IU collaboration. A well-known artificial neural network (ANN) model, was trained and cross-validated to develop the predictive model. Findings The study identified company size (number of employees and approximate annual turnover), the length of experience in the construction industry, previous IU collaboration, the importance of innovation and motivation of innovation for short term showed statistically significant influence on the likelihood of collaboration. The study also revealed there was an increase in interest amongst companies to engage the university in collaborative research. The ANN model successfully predicted the likelihood of a potential construction partner to collaborate with universities at the accuracy of 85.5%, which was considered as a reasonably good model. Originality/value The study investigated the nature of collaboration and the factors that can have an impact on the potential IU collaborations and based on that, introduced the implementation of machine learning approach to examine the likelihood of IU collaboration. While the developed model was derived from analysing data set from Western Australian construction industry, the methodology proposed here can be used as the basis of predictive developing models for construction industry elsewhere to help universities in assessing the likelihood for collaborating and partnering with the targeted construction companies.

scholarly journals A machine learning predictive model of in-hospital mortality in patients with sepsis complicated by anemia: a retrospective study based on the MIMIC-III database

2021 ◽  
Author(s):  
Xiaobin Liu ◽  
Yu Zhao ◽  
Yingyi Qin ◽  
Dan Wang ◽  
Xi Yin ◽  
...  
Keyword(s):  
Machine Learning ◽  
Hospital Mortality ◽  
Predictive Model ◽  
Clinical Data ◽  
Predictive Models ◽  
Goodness Of Fit ◽  
Advantages And Disadvantages ◽  
Extreme Gradient Boosting ◽  
Comorbidity Index ◽  
Mimic Iii

Abstract BackgroudPatients with sepsis complicated by anemia have a higher risk of mortality. It is clinically important to study the risk factors associated with the prognosis of this disease. The aim of this study was to establish a predictive model of mortality during hospitalization by extracting clinical data from the Medical Information Mart for Intensive Care III (MIMIC-III) database. MethodsThe clinical data of patients with sepsis complicated by anemia in the MIMIC-III database were retrospectively analyzed. Indexes were screened by stepwise logistic regression (LR), and machine learning predictive models such as Decision Tree (DT), Random Forests (RF), and eXtreme Gradient Boosting (XGBoost) were developed and compared, identifying advantages and disadvantages of each model. ResultsA total of 13,547 patients with sepsis complicated by anemia were included in the study, among which 1,827 died during hospitalization and 11,720 were still alive at discharge. The preliminary stepwise regression model selected 20 clinical indexes, including Elixhauser comorbidity index, maximum blood urea nitrogen (BUN), and maximum hemoglobin reduction. The predictive models showed good discriminative ability (area under the receiver operating characteristic curve [AUROC]:LR, 0.777; DT, 0.726; RF, 0.788; XGBoost, 0.815) and goodness of fit (area under the precision-recall curve [AUPRC]: LR, 0.350; DT, 0.290; RF, 0.400; XGBoost, 0.428). The Shapley Additive exPlanation (SHAP) values in the XGBoost model showed that Elixhauser comorbidity index, maximum BUN, maximum hemoglobin reduction, ventilator use within 24 hours of admission, and age were significant features for predicting in-hospital mortality in patients with sepsis complicated by anemia. ConclusionsThe XGBoost model had better discrimination ability and goodness of fit when compared with other models. Machine learning algorithms have significant practical value in the development of an early warning system for patients with sepsis complicated by anemia.

Time dependent patient no-show predictive modelling development

2016 ◽  
Vol 29 (4) ◽  
pp. 475-488 ◽  
Author(s):  
Yu-Li Huang ◽  
David A. Hanauer
Keyword(s):  
Predictive Model ◽  
Predictive Models ◽  
Negative Impact ◽  
Area Under The Curve ◽  
Wait Time ◽  
Predictive Modelling ◽  
Time Dependent ◽  
Misclassification Rate ◽  
Data Set ◽  
Content Type

Purpose – The purpose of this paper is to develop evident-based predictive no-show models considering patients’ each past appointment status, a time-dependent component, as an independent predictor to improve predictability. Design/methodology/approach – A ten-year retrospective data set was extracted from a pediatric clinic. It consisted of 7,291 distinct patients who had at least two visits along with their appointment characteristics, patient demographics, and insurance information. Logistic regression was adopted to develop no-show models using two-thirds of the data for training and the remaining data for validation. The no-show threshold was then determined based on minimizing the misclassification of show/no-show assignments. There were a total of 26 predictive model developed based on the number of available past appointments. Simulation was employed to test the effective of each model on costs of patient wait time, physician idle time, and overtime. Findings – The results demonstrated the misclassification rate and the area under the curve of the receiver operating characteristic gradually improved as more appointment history was included until around the 20th predictive model. The overbooking method with no-show predictive models suggested incorporating up to the 16th model and outperformed other overbooking methods by as much as 9.4 per cent in the cost per patient while allowing two additional patients in a clinic day. Research limitations/implications – The challenge now is to actually implement the no-show predictive model systematically to further demonstrate its robustness and simplicity in various scheduling systems. Originality/value – This paper provides examples of how to build the no-show predictive models with time-dependent components to improve the overbooking policy. Accurately identifying scheduled patients’ show/no-show status allows clinics to proactively schedule patients to reduce the negative impact of patient no-shows.

Investigation of the Transient Characteristics for Laser Cladding Beads Using 420 Stainless Steel Powder

2017 ◽  
Vol 139 (8) ◽  
Author(s):  
S. M. Saqib ◽  
R. J. Urbanic
Keyword(s):  
Real Time ◽  
Laser Cladding ◽  
Process Planning ◽  
Control Strategies ◽  
Steel Powder ◽  
Process Models ◽  
Bead Geometry ◽  
Real Time Control ◽  
Transient Conditions ◽  
Simulation Tools

To understand the different aspects of the laser cladding (LC) process, process models can be of aid. Presently, the correct parameter settings for different manufacturing processes, such as machining and casting, are based on simulation tools that can evaluate the influence of the process parameters for different conditions. However, there are no comprehensive, focused simulation process planning tools available for the LC process. In the past, most of the research has focused on the experimentally based optimization strategies for a process configuration, typically for a single track bead in steady-state conditions. However, an understanding of realistic transient conditions needs to be explored for effective process planning simulation tools and build strategies to be developed. A set of cladding experiments have been performed for single and multiple bead scenarios, and the effects of the transient conditions on the bead geometry for these scenarios have been investigated. It is found that the lead-in and lead-out conditions differ, corner geometry influences the bead height, and when changing the input power levels, the geometry values oscillate differently than the input pulses. Changes in the bead geometry are inherent when depositing material; consequently, real-time adjustments for the process setting are essential. The dynamic, time varying heating and solidification, for multiple layer scenarios, leads to challenging process planning and real-time control strategies.

Cold spraying and laser cladding as an alternative to electroplating processes

2019 ◽  
Vol 91 (2) ◽  
pp. 205-215 ◽  
Author(s):  
Jaroslaw Sienicki ◽  
Wojciech Zórawski ◽  
Adam Dworak ◽  
Piotr Koruba ◽  
Piotr Jurewicz ◽  
...  
Keyword(s):  
Cold Spray ◽  
Laser Cladding ◽  
Process Parameters ◽  
Process Development ◽  
Research Work ◽  
Cold Spraying ◽  
High Energy ◽  
Titanium Coating ◽  
Low Carbon ◽  
Content Type

Purpose The purpose of this paper is to propose cold spraying and laser cladding processes as alternatives to cadmium and chromium electroplating, respectively. There are many substances or chemicals within the coating technology that can be identified as substances of very high concern because of their carcinogenic or mutagenic nature. Cadmium and chromium undoubtedly belong to these items and are the basic constituents of electrolytic coating processes. Finding an alternative and adapting to the existing restrictions of the usage of such hazardous products stands for many to be or not to be in the market. Design/methodology/approach The research work was focused on down selecting the appropriate materials, producing the coating samples, testing their properties and optimizing process parameters by statistical method. On the one hand, the high-pressure cold spray system and spraying of the titanium coating on the landing gear component, and on the other hand, the high-energy laser cladding facility and the wear resistant cobalt-based coating deposited onto the shock absorber piston. Substrates of these two applications were made of the same material, 4330 – high-strength low-carbon steel. Findings Meeting the requirements of Registration, Evaluation, Authorization and Restriction of Chemicals implies undertaking research and implementation work to identify alternative processes. The work provides the technical characteristics of new coatings justifying application readiness of the researched processes. Originality/value Taguchi’s design of experiment method was combined with the measurements and analysis of specified coating properties for the optimization of the cold spray process parameters. There is also laser cladding process development presented as a fast rate technology generating coatings with the unique properties.

A fuzzy-regression-PSO based hybrid method for selecting welding conditions in robotic gas metal arc welding

2020 ◽  
Vol 40 (4) ◽  
pp. 601-612
Author(s):  
Amruta Rout ◽  
Deepak Bbvl ◽  
Bibhuti B. Biswal ◽  
Golak Bihari Mahanta
Keyword(s):  
Process Parameters ◽  
Optimization Technique ◽  
Hybrid Approach ◽  
Fuzzy Regression ◽  
Weld Strength ◽  
Optimization Approach ◽  
Bead Geometry ◽  
Weld Quality ◽  
Welding Condition ◽  
Content Type

Purpose This paper aims to propose fuzzy-regression-particle swarm optimization (PSO) based hybrid optimization approach for getting maximum weld quality in terms of weld strength and bead depth of penetration. Design/methodology/approach The prediction of welding quality to achieve best of it is not possible by any single optimization technique. Therefore, fuzzy technique has been applied to predict the weld quality in terms of weld strength and weld bead geometry in combination with a multi-performance characteristic index (MPCI). Then regression analysis has been applied to develop relation between the MPCI output value and the input welding process parameters. Finally, PSO method has been used to get the optimal welding condition by maximizing the MPCI value. Findings The predicted weld quality or the MPCI values in terms of combined weld strength and bead geometry has been found to be highly co-related with the weld process parameters. Therefore, it makes the process easy for setting of weld process parameters for achieving best weld quality, as there is no need to finding the relation for individual weld quality parameter and weld process parameters although they are co-related in a complicated manner. Originality/value In this paper, a new hybrid approach for predicting the weld quality in terms of both mechanical properties and weld geometry and optimizing the same has been proposed. As these parameters are highly correlated and dependent on the weld process parameters the proposed approach can effectively analyzing the ambiguity and significance of each process and performance parameter.

Three-dimensional numerical study of heat-affected zone in induction welding of tubes

2020 ◽  
Vol 39 (1) ◽  
pp. 213-219
Author(s):  
Prerana Das ◽  
John Inge Asperheim ◽  
Bjørnar Grande ◽  
Thomas Petzold ◽  
Dietmar Hömberg
Keyword(s):  
Process Parameters ◽  
Material Properties ◽  
High Frequency ◽  
3D Model ◽  
Welding Process ◽  
Induction Coil ◽  
Content Type ◽  
Non Linear ◽  
Tube Welding ◽  
Welding Line

Purpose Quality of the weld joint produced by high-frequency induction (HFI) welding of steel tubes is attributed to a number of process parameters. There are several important process parameters such as the speed of the welding line, the angle of the approaching strip edges, the physical configuration of the induction coil, impeder, formed steel strip and weld rolls with respect to each other, the pressure of the weld rolls and frequency of the high-frequency current in the induction coil. The purpose of this paper is to develop a 3D model of tube welding process that incorporates realistic material properties and movement of the strip. Design/methodology/approach 3D numerical simulation by the finite element method (FEM) can be used to understand the influence of these process parameters. In this study, the authors have developed a quasi-steady model along with the coupling of electromagnetic and thermal model and incorporation of non-linear electromagnetic and thermal material properties. Findings In this study, 3D FEM model has been established which gives results in accordance with previously published work on induction tube welding. The effect of the Vee-angle and frequency on the temperature profile created in the strip edge during the electromagnetic heating is studied. Practical implications The authors are now able to simulate the induction tube welding process at a more reasonable computational cost enabling an analysis of the process. Originality/value A 3D model has been developed for induction tube welding. A non-linearly coupled system of Maxwell’s electromagnetic equation and the heat equation is implemented using the fixed point iteration method. The model also takes into account non-linear magnetic and thermal material properties. Adaptive remeshing is implemented to optimise mesh size for the electrical skin depth of induced current in the strip. The model also accounts for the high welding-line speeds which influence the mode of heat transfer in the strip.

A model of adoption determinants of ERP within T-O-E framework

2016 ◽  
Vol 29 (4) ◽  
pp. 901-930 ◽  
Author(s):  
Hart O. Awa ◽  
Ojiabo Ukoha Ojiabo
Keyword(s):  
Data Collection ◽  
Enterprise Resource Planning ◽  
Goodness Of Fit ◽  
Resource Planning ◽  
Likelihood Ratios ◽  
Content Type ◽  
Business Operations ◽  
Port Harcourt ◽  
Perceived Values ◽  
Trading Partners

Purpose The purpose of this paper is to attempts to provide further insight into IS adoption by investigating how 12 factors within the technology-organization-environment framework explain small- and medium-sized enterprises’ (SMEs) adoption of enterprise resource planning (ERP) software. Design/methodology/approach The approach for data collection was questionnaire survey involving executives of SMEs drawn from six fast service enterprises with strong operations in Port Harcourt. The mode of sampling was purposive and snow ball and analysis involves logistic regression test; the likelihood ratios, Hosmer and Lemeshow’s goodness of fit, and Nagelkerke’s R2 provided the necessary lenses. Findings The 12 hypothesized relationships were supported with each factor differing in its statistical coefficient and some bearing negative values. ICT infrastructures, technical know-how, perceived compatibility, perceived values, security, and firm’s size were found statistically significant adoption determinants. Although, scope of business operations, trading partners’ readiness, demographic composition, subjective norms, external supports, and competitive pressures were equally critical but their negative coefficients suggest they pose less of an obstacle to adopters than to non-adopters. Thus, adoption of ERP by SMEs is more driven by technological factors than by organizational and environmental factors. Research limitations/implications The study is limited by its scope of data collection and phases, therefore extended data are needed to apply the findings to other sectors/industries and to factor in the implementation and post-adoption phases in order to forge a more integrated and holistic adoption framework. Practical implications The model may be used by IS vendors to make investment decisions, to meet customers’ needs, and to craft informed marketing programs that would appeal to actual and potential adopters and cause them to progress in the customer loyalty ladder. Originality/value The paper contributes to the growing research on IS innovations’ adoption by using factors within the T-O-E framework to explains SMEs’ adoption of ERP.

Study of the route correction system for the on-board navigation system of an unmanned aerial vehicle on the grounds of radar images of the terrain

2020 ◽  
pp. 129-134
Keyword(s):  
Predictive Model ◽  
Unmanned Aerial Vehicle ◽  
Predictive Models ◽  
Navigation System ◽  
Error Compensation ◽  
Compensation Scheme ◽  
Radar Images ◽  
Aerial Vehicle ◽  
System Errors ◽  
Correction System

The system of route correction of an unmanned aerial vehicle (UAV) is considered. For the route correction the on-board radar complex is used. In conditions of active interference, it is impossible to use radar images for the route correction so it is proposed to use the on-board navigation system with algorithmic correction. An error compensation scheme of the navigation system in the output signal using the algorithm for constructing a predictive model of the system errors is applied. The predictive model is building using the genetic algorithm and the method of group accounting of arguments. The quality comparison of the algorithms for constructing predictive models is carried out using mathematical modeling.

Sign in / Sign up

Export Citation Format

Share Document