Prediction Of Key Performance Indicators Research Articles

A Two-Stage Multi-Target Domain Adaptation Framework for Prediction of Key Performance Indicators Based on Adversarial Network

A Two-Stage Multi-Target Domain Adaptation Framework for Prediction of Key Performance Indicators Based on Adversarial Network

An explainable artificial intelligence based approach for the prediction of key performance indicators for 1 megawatt solar plant under local steppe climate conditions

Motivated by the growing significance of solar energy, the paper presents the findings of the maiden study on the long-term seasonal performance assessment of three PV module technologies—monocrystalline (m-Si), polycrystalline (p-Si), and amorphous silicon (a-Si) housed in a 1 MW solar plant installed at a local steppe climate in Gandhinagar, Gujarat, India using the key performance indicators (KPIs), daily power generation, final yield (Yf), reference yield (Yr), total energy loss (TEL) and performance ratio (PR). Notably, the am-Si PV modules were found to be the best performing technology, with an average PR of 71.26% and the lowest average total energy loss of 631 h during the monitored period. Moreover, PV systems have become more popular as a source of green, clean energy; yet, they have low base-load energy sustainability. Explainable AI (XAI) aids in understanding predictions made by the machine learning models. The importance of XAI lies in the comprehension and trust which is embedded into the results of machine learning algorithms. Therefore, the present work builds a comprehensive XAI model for predicting KPIs and assesses the performance of several learning algorithms using R-value, MAE, the number of iterations, and execution time. The Levenberg-Marquardt algorithm predicts the KPIs for p-Si, am-Si, and m-Si with prediction accuracies of 98.63%, 98.58%, and 90.09%, respectively. In summary, this study not only advances our understanding of PV module performance in a steppe climate context but also contributes to the broader field by integrating an XAI model for transparent and reliable predictions of the KPIs.

Artificial Neural Network Based Prediction of Key Performance Indicators for Mobile Telecommunications

This paper presents comparativestudies among artificial neural network neurons. Four key performance indicators were predicted using neural network. The key performance indicators and weather parameters for Osun State University, Osogbo, Nigeria were employed. MATLAB R2020a was employed to develop the neural network models. Three different neural network models were developed. Model A, Model B and Model C with ten neurons, fifteen neurons and twenty neurons respectively, the hidden layer of the models was Log-sigmoid activation function, and the linear activation was used at the output layer of the models. The three models were compared using mean absolute error and mean square error. The best performing model was Model B with fifteen neurons. Its mean absolute error and mean square error is 0.0909 and 0.0123 respectively. Model A with ten neurons was the least performing model with mean absolute error and mean square error of 0.0990 and 0.0148 respectively. The results show that for a model to be robust, several neurons should be tested to establish the most effective model.

A multimode structured prediction model based on dynamic attribution graph attention network for complex industrial processes

Complex industrial processes with dynamic and time-varying characteristics, as well as diverse operating conditions, pose challenges in developing accurate real-time online prediction models for the key performance indicator (KPI). To address these challenges, a KPI prediction framework is proposed in this study, which utilizes a multimode structured prediction model based on a dynamic-attribution graph attention network (DAGAT). First, a multimode clustering method based on a Gaussian mixture model is designed to divide and locate industrial multimode operational data. Second, to address the limitations of traditional deep network models in accurately capturing the complex relationships between process variables and KPI, a graph structure is used to model multi-sensor time series data. Subsequently, a novel DAGAT is proposed to improve node prediction performance and interpretation of industrial graph data. Considering the cumulative effect of large-scale equipment, the DAGAT incorporates knowledge of the industrial process, enabling it to capture the influence of chemical-reaction variables. Simultaneously, the Huber loss function is used to further optimize the prediction results. Moreover, the model can calculate in real-time the contribution of each individual process variable to KPI, providing guidance for optimal control. Finally, the effectiveness and superiority of the proposed method are validated through an industrial case.

Saving energy by maximising daylight and minimising the impact on occupants: An automatic lighting system approach

Electric lighting energy expenses has gained importance in the last decades representing a large part of the global energy expense worldwide. The increase of the proportion of glazed building facades is an incomplete solution raising other problems regarding thermal and visual comfort, while rebound effects in terms of lighting usage lowered the energetic benefits of LED technology. Another strategy to address the issue of energy is the well-known strategy of the re-introduction of task lights. The current study presents a system arbitrating the contribution of daylight with general and task electric lighting from the prediction of key performance indicators (work plane illuminance and daylight glare probability) by a machine learning model. It aims at reducing energy consumption while minimizing the impact on occupants’ comfort and productivity. This system was validated by the mean of a controlled user experiment with 60 participants. Altogether, our data revealed that RADIANCE simulations model accurately predicts on-site monitoring (R2=.96) and that the machine learning model accurately predicts RADIANCE simulations for both work plane illuminance (R2=.99) and daylight glare probability (R2=.99) allowing to operate the automated system. Data from the user experiment shows that our system successfully saves 35 Wh on average compared to manual operation of the room by users, by making prior use of daylight and task lighting. This represents more than halving the energy demand. Importantly, the automation impacted neither users’ comfort or task performance. System acceptance constructs such as system reliability or trust in the system were found to be linked with an attribution of the locus of control to powerful others. This result suggests that the key towards high user acceptance might lie in purposefully addressing this dimension. Altogether, the presented results support the potential for energy saving through the development of user-centred lighting automation.

Explainable AI for Solar Energy: A Deep Learning-Based Approach for the Prediction of Key Performance Indicators for 1 MW Solar Plant Under Local Steppe Climate Conditions

Explainable AI for Solar Energy: A Deep Learning-Based Approach for the Prediction of Key Performance Indicators for 1 MW Solar Plant Under Local Steppe Climate Conditions

Improved Ensemble Feature Selection Based on DT for KPI Prediction

In the production process of large-scale machinery and complex industries, the key performance indicator (KPI) prediction is an essential part of project scheduling and cost estimation. The continuous enrichment of sensor types and functions brings us massive soft-sensing parameters for regression, but also brings severe challenges to algorithm learning. In this paper, an improved ensemble feature selection based on decision tree (EFS-DT) strategy for KPI prediction is developed. On the one hand, the ensemble of multi-criteria filtering results broadens the selector’s perspective without the time cost of superposition. On the other hand, credibility and similarity analysis are designed to eliminate the concerns of Dempster’s combination rule about conflict. After re-evaluating the variable scores, more high-quality variables can be selected to build a more accurate and robust KPI prediction model. Finally, a realistic shield tunnel case in China is used to evaluate the feasibility and effectiveness of proposed approach.

Deep Learning-Based Prediction of Key Performance Indicators for Electrical Machines

The design of an electrical machine can be quantified and evaluated by Key Performance Indicators (KPIs) such as maximum torque, critical field strength, costs of active parts, sound power, etc. Generally, cross-domain tool-chains are used to optimize all the KPIs from different domains (multi-objective optimization) by varying the given input parameters in the largest possible design space. This optimization process involves magneto-static finite element simulation to obtain these decisive KPIs. It makes the whole process a vehemently time-consuming computational task that counts on the availability of resources with the involvement of high computational cost. In this paper, a data-aided, deep learning-based meta-model is employed to predict the KPIs of an electrical machine quickly and with high accuracy to accelerate the full optimization process and reduce its computational costs. The focus is on analyzing various forms of input data that serve as a geometry representation of the machine. Namely, these are the cross-section image of the electrical machine that allows a very general description of the geometry relating to different topologies and the classical way with scalar parametrization of geometry. The impact of the resolution of the image is studied in detail. The results show a high prediction accuracy and proof that the validity of a deep learning-based meta-model to minimize the optimization time. The results also indicate that the prediction quality of an image-based approach can be made comparable to the classical way based on scalar parameters.

A Time Series Prediction Approach Based on Hybrid Tuning for Database Performance Indicator in AIOps

One of the most important applications of the intelligent operation and maintenance of a cloud database is its trend prediction of key performance indicators (KPI), such as disk use, memory use, etc. We propose a method named AutoPA4DB (Auto Prophet and ARIMA for Database) to predict the trend of the KPIs of the cloud database based on the Prophet model and the ARIMA model. Our AutoPA4DB method includes data preprocessing, model building, parameter tuning and optimization. We employ the weighted MAPE coverage to measure its accuracy and use 6 industrial datasets including 10 KPIs to compare the AutoPA4DB method with other three time-series trend prediction algorithms. The experimental results show that our AutoPA4DB method performs best in predicting monotonic variation data, e.g.disk use trend prediction. But it is unstable in predicting oscillatory variation data; for example, it is acceptable in memory use trend prediction but has poor accuracy in predicting the number of database connection trends.

Deep Learning for Industrial KPI Prediction: When Ensemble Learning Meets Semi-Supervised Data

Soft-sensing techniques are of great significance in industrial processes for monitoring and prediction of key performance indicators. Due to the effectiveness of nonlinear feature extraction and strong expansibility, an autoencoder (AE) and its extensions have been widely developed for industrial applications. Nevertheless, an AE commonly uses the last hidden layer for regression modeling with the output, which seems to be a kind of information waste as the shallow layers are also abstractions of input data. Besides, when there are excessive unlabeled samples, AE-based models are less likely to make full use of them or even degrade the performance. To deal with these issues, a method called ensemble semi-supervised gated stacked AE (ES <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> GSAE) is proposed in this article. Gate units are used to develop connections between different layers and the output layer, which also help quantify the contribution of different hidden layers. Moreover, the idea of ensemble learning is combined with semi-supervised learning, in which different unlabeled datasets are used for training different submodels to ensure their diversities. In this way, unlabeled samples can be utilized more efficiently and help enhance the model performance. The effectiveness and superiority are verified in a real industrial process by comparing the proposed method with other typical AE-based models.

Novel method using Dubinin-Astakhov theory in sorption reactor design for refrigeration and heat pump applications

This paper presents new methodology that is based on a single given adsorbent-refrigerant pair characteristic (such as activated carbon 208C) leading to the characterization of the same adsorbent (activated carbon 208C) with various refrigerants such as Water, Methanol, Ethanol, R723 (azeotropic mixture of 60% Ammonia and 40% Dimethyl Ether), Dimethyl Ether (DME) and Carbon Dioxide (R744). Overall, the results obtained with both Methanol and Carbon Dioxide (R744) show that the new method predicts the refrigerant uptake with a marginal difference (less than 5%) compared to standard method that heavily depends on experimental data. For example with methanol, the standard method produces a maximum uptake (xo) of 0.3676kg methanol/kg carbon while the new method predicts 0.3740kg methanol/kg carbon; with CO2 both standard and new methods predict 0.3242kg CO2/kg carbon and 0.3190kg CO2/kg carbon respectively. The results exploitation of this method led to rapid prediction of key performance indicators of adsorption system utilizing compacted activated carbon 208C-R723 refrigerant pair for ice making, air conditioning and heat pump applications.

An Improved Incremental Learning Approach for KPI Prognosis of Dynamic Fuel Cell System.

The key performance indicator (KPI) has an important practical value with respect to the product quality and economic benefits for modern industry. To cope with the KPI prognosis issue under nonlinear conditions, this paper presents an improved incremental learning approach based on available process measurements. The proposed approach takes advantage of the algorithm overlapping of locally weighted projection regression (LWPR) and partial least squares (PLS), implementing the PLS-based prognosis in each locally linear model produced by the incremental learning process of LWPR. The global prognosis results including KPI prediction and process monitoring are obtained from the corresponding normalized weighted means of all the local models. The statistical indicators for prognosis are enhanced as well by the design of novel KPI-related and KPI-unrelated statistics with suitable control limits for non-Gaussian data. For application-oriented purpose, the process measurements from real datasets of a proton exchange membrane fuel cell system are employed to demonstrate the effectiveness of KPI prognosis. The proposed approach is finally extended to a long-term voltage prediction for potential reference of further fuel cell applications.

Trading Agents’ Negotiation in Business Management Using Demand Functions: Simulation Experiments with Binomial Distribution

The aim of this paper is to propose an experimental business management approach to cover a seller-to-customer price negotiation in an agent-based simulations. The core element in this approach is the price negotiation. We used Marshallian demand function and a Cobb-Douglas utility function in the negotiation process. Moreover, multi-agent model is proposed and implemented in Jade development platform. Its task is to serve as a simulation framework for the trading processes execution. The main background of this framework is to be integrated in management information systems as a decision support module for a prediction of key performance indicators of a virtual company. A binomial distribution was used in presented experiments to simulate the quantity of negotiated commodities. The paper firstly presents some of the existing principles about consumer behavior, agent-based modeling and simulation in the same area and demand function theory. Secondly, presents multi-agent model and demand functions negotiations more formally. Finally, shows some of the simulation results in a trading processes throughout one year of selling commodities to consumers. The results obtained show that the demand functions could be properly used to simulate trading processes.

A Novel Scheme for Key Performance Indicator Prediction and Diagnosis With Application to an Industrial Hot Strip Mill

In this paper, a data-driven scheme of key performance indicator (KPI) prediction and diagnosis is developed for complex industrial processes. For static processes, a KPI prediction and diagnosis approach is proposed in order to improve the prediction performance. In comparison with the standard partial least squares (PLS) method, the alternative approach significantly simplifies the computation procedure. By means of a data-driven realization of the so-called left coprime factorization (LCF) of a process, efficient KPI prediction, and diagnosis algorithms are developed for dynamic processes, respectively, with and without measurable KPIs. The proposed KPI prediction and diagnosis scheme is finally applied to an industrial hot strip mill, and the results demonstrate the effectiveness of the proposed scheme.