Estimation Of Key Performance Indicators Research Articles

Automated machine learning methodology for optimizing production processes in small and medium-sized enterprises

Machine learning can be effectively used to generate models capable of representing the dynamic of production processes of small and medium-sized enterprises. These models enable the estimation of key performance indicators, and are often used for optimizing production processes. However, in most industrial applications, modeling and optimization of production processes are currently carried out as separate tasks, manually in a very costly and inefficient way. Automated machine learning tools and frameworks facilitate the path for deriving models, reducing modeling time and cost. However, optimization by exploiting production models is still in infancy. This work presents a methodology for integrating a fully automated procedure that embraces automated machine learning pipelines and a multi-objective optimization algorithm for improving the production processes, with special focus on small and medium-sized enterprises. This procedure is supported on embedding the generated models as objective functions of a reference point based non-dominated sorting genetic algorithm, resulting in preference-based Pareto-optimal parametrizations of the corresponding production processes. The methodology was implemented and validated using data from a manufacturing production process of a small manufacturing enterprise, generating highly accurate machine learning-based models for the analyzed indicators. Additionally, by applying the optimization step of the proposed methodology it was possible to increase the productivity of the manufacturing process by 3.19 % and reduce its defect rate by 2.15 %, outperforming the results obtained with traditional trial and error method focused on productivity alone.

Cost-Based Decision Support System: A Dynamic Cost Estimation of Key Performance Indicators in Manufacturing

An attempt is made to translate five generic key performance indicators (KPIs) into a continuous real-time cost function in a batch order-based manufacturing environment. The challenge of controlling and optimizing resource utilization, production efficiency, product-process quality, environmental impact, and inventory was specified by microelectronics and hard metal composite manufacturers. The motivation is to facilitate decision-making by converting operations management data into dynamic financial cost models. The process of interpreting engineering data of the physical level and operations management level into financial metrics creates a common language between engineers, managers, and financial departments of the company whose common objective is the profitability of the company, each with their own priorities. The proposed method provides a realistic representation of the performance of the system in monetary value. The integration may become an instrument of effective and efficient tactical and strategic collective decision-making. The main outcome is a near real-time formulation and prediction of manufacturing cost with respect to the five KPIs. The resultant cost function is verified according to several production scenarios. The case study demonstrating the proposed cost modeling methodology utilizes real-time and historical information from two different industrial partners in Tungsten metallurgy and electronic circuit manufacturing industries.

A spatiotemporal‐distributed deep‐learning framework for KPI estimation of chemical processes with cascaded reactors

AbstractOnline and accurate estimation of key performance indicators (KPI) is the foundation for operational optimization of a chemical process. However, a chemical process usually consists of multiple reactors, and the factors influencing KPI are spatially distributed in the long process flow. In addition, due to the distinct time lags between KPI and each reactor, temporal relationships among KPI and its influence factors are a mixture of short‐term and long‐term relationships. In this regard, a deep distributed KPI estimator with a self‐attention mechanism is proposed in this paper. First, considering the process topology, a cascaded long short‐term memory network is developed to simulate the process topology and capture the short‐term effects. Then, to extract the long‐term dependencies, a de‐noise self‐attention layer is employed to model interactions of all the influence factors explicitly and dynamically. Lastly, the proposed method is compared with typical state‐of‐the‐art methods using real industrial data. The comparison results illustrate the performance and effectiveness of the proposed KPI estimation method.

Impact of User Playback Interactions on In-Network Estimation of Video Streaming Performance

With significant growth in video streaming services, coupled with widespread use of traffic encryption, network operators are faced with the challenge of monitoring Key Performance Indicators (KPI) needed to detect quality impairments and drive Quality of Experience (QoE) management mechanisms. Though literature suggests that QoE/KPIs can be inferred from encrypted network traffic using machine learning (ML) methods, most studies published thus far fail to account for frequent viewer interactions, such as pauses, seeking forward/backward, or video abandonment. Such playback-related interactions inherently impact traffic patterns used as input for ML–based KPI estimation models. In this paper, we investigate to what extent network operators can monitor application-layer KPIs considering realistic user interactions, focusing on the use-case of a popular streaming platform with videos streamed to mobile devices. We first investigate the impact of user interactions (pause, seek forward, and abandonment) on the performance of both session-based and real-time KPI classification models trained on datasets that do not contain interactions. Secondly, we systematically evaluate the performance of KPI estimation models trained on datasets including specific sets of interactions to determine which types of interactions need to be included in the model training procedure in order to be applicable for realistic streaming sessions.

Advanced and Complex Energy Systems Monitoring and Control: A Review on Available Technologies and Their Application Criteria.

Complex energy monitoring and control systems have been widely studied as the related topics include different approaches, advanced sensors, and technologies applied to a strongly varying amount of application fields. This paper is a systematic review of what has been done regarding energy metering system issues about (i) sensors, (ii) the choice of their technology and their characterization depending on the application fields, (iii) advanced measurement approaches and methodologies, and (iv) the setup of energy Key Performance Indicators (KPIs). The paper provides models about KPI estimation, by highlighting design criteria of complex energy networks. The proposed study is carried out to give useful elements to build models and to simulate in detail energy systems for performance prediction purposes. Some examples of energy complex KPIs based on the integration of the Artificial Intelligence (AI) concept and on basic KPIs or variables are provided in order to define innovative formulation criteria depending on the application field. The proposed examples highlight how modeling a complex KPI as a function of basic variables or KPIs is possible, by means of graph models of architectures.

A Model-Based Approach Towards Real-Time Analytics in NFV Infrastructures

Network Functions Virtualization (NFV) has recently gained much popularity in the research scene for the flexibility and programmability that it will bring with the software implementation of network functions on commercial off-the-shelf (COTS) hardware. To substantiate its roll out, a number of issues (e.g., COTS’ inherent performance and energy efficiency, virtualization overhead, etc.) must be addressed, in a scalable and sustainable manner. Numerous works in the scientific literature manifest the strong correlation of network key performance indicators (KPIs) with the burstiness of the traffic. This paper proposes a novel model-based analytics approach for profiling virtualized network function (VNF) workloads, towards real-time estimation of network KPIs (specifically, power and latency), based on an MX/G/1/SET queueing model that captures both the workload burstiness and system setup times (caused by interrupt coalescing and power management actions). Experimental results show good estimation accuracies for both VNF workload profiling and network KPI estimation, with respect to the input traffic and actual measurements, respectively. This demonstrates that the proposed approach can be a powerful tool for scalable and sustainable network/service management and orchestration.

A Framework for in-Network QoE Monitoring of Encrypted Video Streaming

With the amount of global network traffic steadily increasing, mainly due to video streaming services, network operators are faced with the challenge of efficiently managing their resources while meeting customer demands and expectations. A prerequisite for such Quality-of-Experience-driven (QoE) network traffic management is the monitoring and inference of application-level performance in terms of video Key Performance Indicators (KPIs) that directly influence end-user QoE. Given the persistent adoption of end-to-end encryption, operators lack direct insights into video quality metrics such as start-up delays, resolutions, or stalling events, which are needed to adequately estimate QoE and drive resource management decisions. Numerous solutions have been proposed to tackle this challenge on individual use-cases, most of them relying on machine learning (ML) for inferring KPIs from observable traffic patterns and statistics. In this paper, we summarize the key findings in state-of-the-art research on the topic. Going beyond previous work, we devise the concept of a generic framework for ML-based QoE/KPI monitoring of HTTP adaptive streaming (HAS) services, including model training, deployment, and re- evaluation. Components of the framework are designed in a generic way, independent of a particular streaming service and platform. The methodology for applying different framework components is discussed across various use-cases. In particular, we demonstrate framework applicability in a concrete use-case involving the YouTube service delivered to smartphones via the mobile YouTube app, as this presents one of the most prominent examples of accessing YouTube. We tackle both QoE/KPI estimation on a per-video-session level (utilizing the validated ITU-T P.1203 QoE model), as well as “real-time” KPI estimation over short time intervals. Obtained results provide important insights and challenges related to the deployment of a generic in-network QoE monitoring framework for encrypted video streams.

Modeling Framework for Comparing Taxi Operational Modes: Case Study in Barcelona

Abstract This paper presents an aggregated mathematical model for the estimation of key performance indicators of the taxi market based on the system’s generalized cost function, which is calculated using the expected statistical values of customers’ trip distance, waiting/access time and the cost of the involved actors, including externalities, who are the taxi drivers, the taxi customers and the city represented by the rest of the drivers and the citizens. Optimum values for the taxi supply are obtained from mathematical formulations depending on the demand level and the size of the city. The model is developed for stand, hailing and dispatching taxi markets and the results are compared, presenting conclusions for the best type of market for each demand level and city size. The model is applied in the city of Barcelona, presenting useful conclusions on the performance indicators of the taxi services and the impact of the applied policies as well as the optimum number of taxis for each operational mode, ranging between 30 and 40 vehicles per hour and km2.

Model-based corridor performance analysis – An application to a European case

The paper proposes a methodology for freight corridor performance monitoring that is suitable for sustainability assessments. The methodology, initiated by the EU-funded project SuperGreen, involves the periodic monitoring of a standard set of transport chains along the corridor in relation to a number of Key Performance Indicators (KPIs). It consists of decomposing the corridor into transport chains, selecting a sample of typical chains, assessing these chains through a set of KPIs, and then aggregating the chain-level KPIs to corridor-level ones using proper weights. A critical feature of this methodology concerns the selection of the sample chains and the calculation of the corresponding weights. After several rounds of development, the proposed methodology suggests a combined approach involving the use of a transport model for sample construction and weight calculation followed by stakeholder refinement and verification. The sample construction part of the methodology was tested on GreCOR, a green corridor project in the North Sea Region, using the Danish National Traffic Model as the principal source of information for both sample construction and KPI estimation. The results show that, to the extent covered by the GreCOR application, the proposed methodology can effectively assess the performance of a freight transport corridor. Combining the model-based approach for the sample construction with the study-based approach for the estimation of chain-level indicators exploits the strengths of each method and avoids their weaknesses. Possible improvements are also suggested by the paper.

Современные принципы работы системы управления персоналом

This work describes basic principles of human resource (HR) management improvement by using new information systems and monitoring of key performance indicators (KPI). HR management systems are based on software uniting information from various applications to the one universal database. Relation between HR and competence modules via single database is the main difference from custom and proprietary previous developments that makes this software flexible, while KPI estimation leads to detection of potential improvements. Proposed approach differs from known solutions by containing all main processes of HR management. Earlier on developed systems were characterized by narrow only one goal concerned with as example payroll preparation optimization or work time management. Modern systems cover all problems relating with HR departments including monitoring and improvement of process effectiveness, organization hierarchy management and simplification of all type financial transactions. This work presents researches mainly focused on improvement of large enterprise processes as well as it may be applied for small and medium-sized businesses. Here modern HR management systems are considered under KPI monitoring to estimate a progress in toward strategic goals.

Estimation and monitoring of key performance indicators of manufacturing systems using the multi-output Gaussian process

Recently, the estimation and monitoring of manufacturing key performance indicators (KPIs) have drawn significant attention. In this article, a KPI estimation and monitoring method using a multi-output Gaussian process (MGP) is proposed. The Gaussian process (GP) is an effective non-parametric flexible tool for data-driven statistical modelling for various systems. The unique features of the proposed method is that the MGP enjoys the high flexibility and desirable analytical properties of the GP while also capturing the correlation between different KPIs, thus providing better estimation accuracy and error quantification. The advantageous features of the proposed method are demonstrated through a numerical study as well as a case study with real world data in the estimation and monitoring of throughput for a multiclass production operation.

Machine learning-based dynamic frequency and bandwidth allocation in self-organized LTE dense small cell deployments

Self-organizing networks (SONs) are expected to minimize operational and capital expenditure of the operators while improving the end users’ quality of experience. To achieve these goals, the SON solutions are expected to learn from the environment and to be able to dynamically adapt to it. In this work, we propose a learning-based approach for self-optimization in SON deployments. In the proposed approach, the learning capability has the central role to perform the estimation of key performance indicators (KPIs) which are then exploited for the selection of the optimal network configuration. We apply this approach to the use case of dynamic frequency and bandwidth assignments (DFBA) in long-term evolution (LTE) residential small cell network deployments. For the implementation of the learning capability and the estimation of KPIs, we select and investigate various machine learning and statistical regression techniques. We provide a comprehensive analysis and comparison of these techniques evaluating the different factors that can influence the accuracy of the KPI predictions and consequently the performance of the network. Finally, we evaluate the performance of learning-based DFBA solution and compare it with the legacy approach and against an optimal exhaustive search for best configuration. The results show that the learning-based DFBA achieves on average a performance improvement of 33 % over approaches that are based on analytical models, reaching 95 % of the optimal network performance while leveraging just a small number of network measurements.