Customer Profiles Research Articles

Net Load Forecasting With Disaggregated Behind-the-Meter PV Generation

As worldwide use of residential photovoltaic (PV) systems grows, system operators and utilities will need to transition from forecasting pure demand to forecasting net load with behind-the-meter (BTM) PV generation. However, PV generation can be difficult to predict and the measurements of PV generation from BTM residential systems are often invisible behind a measurement of the net load, making net load forecasting challenging. This paper proposes a novel two-stage framework for net load forecasting in areas with limited observability and high BTM PV generation. First, the profiles of observable customers are used to disaggregate the net load measurements into the pure load and PV generation. Then, separate models are used to forecast the PV generation and pure load individually, and the results are combined for a net load forecast. This paper also proposes a compensator for correcting the error of the net load forecast, using historical forecast errors of the PV generation, pure load, and net load. The proposed framework is tested through two case studies for areas with high BTM PV penetration and less than 10% observable customers. The two-stage forecasting model is compared to two benchmark methods – a time series forecasting model, and a model that forecasts the net load directly using historical net load measurements. Results show that the proposed disaggregation-forecasting framework reduces the error of the net load forecast compared to both benchmark models. In addition, when the net load forecast error is periodic, the compensator can correct the error to improve the forecast accuracy.

Identifying hybrid heating systems in the residential sector from smart meter data

In this paper, we identify hybrid heating systems on a single residential customer’s premises using smart meter data. A comprehensive methodology is developed at a generic level for residential sector buildings to identify the type of primary and support heating systems. The methodology includes the use of unsupervised and supervised learning algorithms both separately and combined. It is applied to two datasets that vary in size, quality of data, and availability and reliability of background information. The datasets contain hourly electricity consumption profiles of residential customers together with the outdoor temperature. The validation metrics for the developed algorithms are elaborated to provide a probabilistic evaluation of the model. The results show that it is possible to identify the types of both primary and support heating systems in the form of probability of having electric- or non-electric type of heating. The results obtained help estimate the flexibility domain of the residential building sector and thereby generate a high value for the energy system as a whole.

A data-driven analytic approach for investigation of electricity demand variability for energy conservation programs

Targeting and designing effective demand response and energy efficiency strategies is a challenging task owing to the variations in electricity usage patterns. To address this problem, we attempt to propose a clustering and Kullback-Leibler (KL) Divergence-based solution to understand the variability across the different consumption patterns. The proposed unsupervised approach utilized hierarchical clustering technique for segmenting the customers based on the similarity in their consumption behaviour, followed by the incorporation of the KL divergence to quantify the variability for different features of customer's load profiles. These features aid in understanding the diversity in load shapes. Depending on the different features, the results reveal that a few of the consumers having certain distinctive consumption can be easily considered as potential for demand flexibility, while others require careful consideration for program offerings. The result also shows significantly higher day-to-day variations in the clustered groups and can be managed by designing suitable behavioural modification strategies. The variability assessment of individual customer profiles helps better comprehend the diversity in usage behaviour. It leads to informed decision-making in planning demand response and energy efficiency initiatives, which was not taken into consideration in the previous studies.

Resource flows in a multi-directional integrated value creation model

This research reveals a multi-directional value creation model that highlights the role of customers as partners in a network of reciprocal resource exchanges. Customer-partners are conceptualized as customers that go beyond purchasing and using a product or service. They voluntarily engage end-customers to help the end-customers use the firms’ products and services, or to otherwise solve the end-customer’s problems with the firms’ products or services. To illustrate the importance of the customer-partner to the value creation process, the authors integrate various value creation perspectives with social and resource exchange theories to illuminate the various resources exchanged when firms and their various types of customers engage in social interactions. The resultant model is derived from a netnography in the context of a high-tech firm’s online support community. Findings point to three important activities (platform design, customer profiling, and platform engagement) for firms to consider if they hope to successfully integrate their customers in the value creation process.

ANALISIS MODEL BISNIS STARTUP KEENANSUPERCAR DENGAN METODE VALUE PROPOSITION DESIGN

Keenansupercar startup is a luxury car rental service platform in Bali. At this time, startup keenansupercar wants to innovate its business model because there is no customer profile and proper product development that matches the wishes of the market/customers. The goal of business model innovation is to create a value proposition for startups, customers, and society. There are various methods that are often used by startup founders in analyzing business models, one of the methods that can be used to analyze business models is value proposition design. The value proposition design method directs startup founders how to use the value proposition canvas in designing and testing a large value proposition on a continuous search for what customers want which will result in a value proposition canvas.

Development of a correlator for flow measurement in pipelines using gamma radiation and cross-correlation function

Measuring water consumption by different users (industrial, agricultural, urban) is essential, since in many countries water royalties per consumer are differentiated by the customer's profile. Therefore, the aim of this study was to develop an alternative flow rate measurement method with low uncertainty for low sampling frequency systems using radiotracers and the Cross-Correlation function. The experimental tests were carried out on a closed water transport line built in Polyvinyl Chloride with 30 m length. The radiotracer used was Na82Br injected as a rapid pulse with volume of 3.0 ml (3 MBq) per test. Data acquisitions were made with a sampling rate ranging between 20 Hz and 1 kHz. A measurement correlator (hardware and software) was developed using special mathematical methods, such as: Savitzky-Golay filtering and Lagrange interpolation to reduce the uncertainty of the flow rate measurement. The correlator was tested in a water flow rate range of 200 L/h to 800 L/h with an uncertainty of 2.0% at a flow rate of 200 L/h.

Does Loyalty Program Membership Lead to Repeat Purchase: A Demographic Analysis with Customer Profiling and Machine Learning

Does Loyalty Program Membership Lead to Repeat Purchase: A Demographic Analysis with Customer Profiling and Machine Learning

Development of a jeans sizing system for young black pear-shaped South African women

The South African Pear-shaped Black women’s jeans market has been confronted by fit dissatisfaction, although there is a growing demand for jeans. This study was approached from the viewpoint that jeans do not fit a large population of this consumer group and investigated the issues related to this problem. Local manufacturers and retailers currently use an adaptation of the British sizing system to suit their customer profile. Not much research, however, has been carried out on developing a sizing system for the pear-shaped figure type, which makes up a sizeable portion of the population in the country. The aim of the study was to establish key variables for developing a jeans sizing system for Black South African pear-shaped women. A total of 60 Black women aged 18 to 35 years at the Cape Peninsula University of Technology were scanned for body measurements using a 3D body scanner. Body measurement differences were examined by using a quantitative research approach to establish the difference between waist and hip measurements. The findings revealed an average drop value of 39cm between waist and hip circumference for a Black pear-shaped figure, in comparison to a drop value of 24cm for a standardised sizing used by the clothing industry. It is recommended that the major stakeholders consider conducting a national anthropometric study to update sizing systems, by using 3D body scanning technology, which provides accurate and consistent measurements of the human body.

Corporate digital responsibility (CDR) in the age of AI: implications for interactive marketing

PurposeDespite all the recent achievements in the field of interactive marketing and artificial intelligence (AI), it is important to consider the ethical implications of these technologies. This paper explains the concept of corporate digital responsibility (CDR) and how it is affected by new advances in AI.Design/methodology/approachThe authors build on the work of Wirtz et al., (2023) and derive several managerial implications for the challenges that AI poses to CDR. CDR refers to a service company's ethical and fair use of data and technology within its digital service ecosystem. It involves establishing standards, protecting customer privacy, conducting external audits and striving for an equitable power dynamic between service firms and their partners.FindingsDespite the risks involved, many companies are not prioritizing good CDR practices. Financial benefits from the collection and use of consumer data, improved customer experience through AI-driven customization and personalization, cost reduction through service automation and the trade-offs between organizational goals and CDR practices can prevent companies from prioritizing good CDR practices.Originality/valueThis is one of the first articles in the service domain to take the concept of CDR and apply it to recent developments in generative AI.Research limitations/implicationsThe emergence of powerful AI tools presents opportunities and challenges. Research opportunities include responsible business restructuring, responsible service automation to ensure fairness and human oversight, addressing dehumanization of service delivery, responsible customer profiling to address privacy and discrimination concerns and preventing AI misuse.

Implementasi Artificial Intelligence (AI) di Bidang Pendidikan Kejuruan Pada Era Revolusi Industri 4.0

Artificial Intelligence (AI) is a concept closely related to technological advancements in the era of Industry 4.0 and Society 5.0. AI can be described as a computer program that involves integrated machine learning, hardware, and software components. This concept of AI is inspired by the neocognitron patterns in the human brain and utilizes knowledge from hardware and software to fulfill its functions. With the progress of Industry 4.0, AI has been widely employed in various fields, including vocational education. The aim of this research is to elucidate the role of AI in vocational education during the Industry 4.0 era, utilizing qualitative research methods. In the context of vocational education, AI is used to personalize learning for each learner. AI systems assist in creating customized learning profiles, such as practical exercises, for individual learners. This enables the adaptation of learning materials to align with the abilities, learning styles, and experiences of each learner. The implementation of AI technology in vocational education is considered to be helpful and facilitates the tasks of educators, particularly in administrative aspects such as assessment and determining final grades based on applicable weights and evaluations. Additionally, AI technology can also create more active and interactive learning experiences, thereby facilitating the teaching and learning process for both educators and learners to achieve their goals.

Segmenting the Malaysian female young adolescent market based on gender roles and identities

PurposeThis paper aims to understand the importance of young female consumers in Asia as a market and propose strategies for retail marketers to effectively engage with them while respecting their evolving gender roles and identities.Design/methodology/approachData are derived from self-administered questionnaires completed by Malaysian female secondary school students aged 13 to 19. Using statements about gender roles and identities, the authors first segment respondents based on their perceptions of ideal female gender roles and identities and subsequently investigate each segment’s shopping orientation.FindingsThis study identifies four segments of young female consumers based on gender identity: civic intellectuals, presentable intellectuals, career-oriented and family-oriented. While young female young consumers with higher presentable intellectual traits are more likely to go window shopping (recreational shoppers), those with higher civic intellectual traits are more inclined to look for things they wish to buy and spend less time doing so (utilitarian shoppers).Practical implicationsMarketers and retailers should continue to use profiling of young female customers to predict the psychological changes related to societal and economic changes. Merely focusing on females’ appearance and family responsibilities may no longer be relevant due to the evolving social order. This implication is especially critical for marketers and retailers targeting the female market.Originality/valueThe results of this study will have important theoretical and managerial implications for marketers and retailers interested in understanding the changing needs and beliefs among young female consumers and how these impacts their shopping orientation.

Utilization of Circulating Tumor DNA in the Surveillance Setting.

National guidelines give recommendations regarding cancer surveillance follow-up. In many early staged cancers radiographic imaging and labs are not routinely recommended unless patients are symptomatic. This can cause a gap in care because commonly when patients present symptomatically, they have progressed and transitioned to later-stage cancer. This study demonstrates how circulating tumor DNA (ctDNA) can be used alongside current guidelines to help screen patients for recurrence in the surveillance setting. A retrospective chart review was performed. Fifty-five charts were reviewed of patients who received ctDNA testing drawn in follow-up after their primary tumor or metastatic disease was rendered surgically or radiographically disease-free. A customized signature profile, using the sixteen most prevalent genomic markers from a patient's primary tumor or biopsy, is developed by whole-exome sequencing. Serial blood draws are then drawn to assess for specific DNA markers using polymerase chain reaction (PCR) assays. Fifty-five charts were reviewed in patients who had stage I-III breast, pancreatic, melanoma, and colorectal cancer. Of the fifty-five, a total of seven had a positive test. Of the seven positive tests, six patients were found to have recurrent/metastatic disease. One positive test was performed four weeks postoperatively but by the second draw ten weeks postoperatively had non-detectable ctDNA. The remaining forty-eight patients had non-detectable ctDNA levels and to date have not had any evidence of recurrence based on standard follow-up guidelines. The utilization of ctDNA in the surveillance setting can be used to help detect recurrence in the surveillance setting.

Comparison of the DBSCAN Algorithm and Affinity Propagation on Business Incubator Tenant Customer Segmentation

The increasingly complex business environment necessitates businesses to design more effective and efficient strategies for company development, including market expansion. To understand customer behaviors, customer data analysis becomes crucial. One common approach used to group customers is segmentation based on RFM analysis (Recency, Frequency, and Monetary). This study aims to compare the performance of two clustering algorithms, namely DBSCAN and Affinity Propagation (AP), in providing customer profile segment recommendations using RFM analysis. DBSCAN algorithm is employed due to its ability to identify arbitrarily shaped clusters and handle data noise. On the other hand, Affinity Propagation (AP) algorithm is chosen for its capability to discover cluster centers without requiring a pre-defined number of clusters. The transaction dataset used in this research is obtained from one of the business incubator tenants at STIKOM Bali. The dataset undergoes preprocessing steps before being segmented using both DBSCAN and AP algorithms. Performance evaluation of the algorithms is conducted using the Silhouette Scores and Davies-Bouldin Index (DBI) matrices. The research findings indicate that the AP algorithm outperforms DBSCAN in this customer segmentation case. The AP algorithm yields Silhouette Scores of 0.699 and DBI of 0.429, along with recommendations for 4 customer segments. Furthermore, further analysis is performed on the AP results using a statistical approach based on the mean values of each segment for the RFM variables. The four customer segments generated by the AP algorithm, based on the mean values of the RFM variables, can be associated with the concept of customer relationship management.

Markov Analysis in Commitment Prediction of Insurance Customer to Minimizing Customer Attrition

Trust and commitment are the important things for insurance customer to continue doing business with insurance company. Customer commitment can be seen from payment history, whether customer doing payment on time or late. This research using Markov Analysis to predict customer commitment using data of insurance premium payment history from each period, identify customer profile, and analyze the type of insurance purchased. This research shows there are three steps of customer who vulnerable to terminate the insurance contract with company. Insurance company could use the result to minimize customer attrition level with several ways like review the applicant profile before make insurance contract, recommends which type of insurance will fit to applicant, and monitor customer who late to do premium payment for two consecutive periods. This research also propose Customer Relationship Management (CRM) strategy to handle customer who match with the three steps of terminating the insurance contract with providing special treatment and service to prevent discontinuance of insurance contract.

Analisa Penerapan Metode Naïve Bayes Dalam Memprediksi Kelayakan Calon Nasabah Dalam Melakukan Pinjaman

Financing through loans is common in the financial world, but the loan approval process requires careful assessment to minimize credit risk that may arise. In this study, we propose the application of the Naïve Bayes method as a tool to predict the eligibility of prospective customers in making loans. This research uses historical datasets from banks that include credit information, payment history, and customer profiles. The analysis process begins with data pre-processing to clean and address missing or incomplete data. Next, the extraction of relevant features and selection of variables are carried out to build a prediction model. The Naïve Bayes method was chosen because of its simple and fast nature in classifying data. This model makes use of the assumption of feature independence, which is useful when feature dimensions are high. In the training phase, the Naïve Bayes model will learn from historical datasets to recognize good and bad customer eligibility patterns. The results of this analysis show that the Naïve Bayes method has succeeded in predicting the eligibility of prospective customers with a satisfactory level of accuracy. In addition, the model is able to identify key factors that affect loan eligibility, such as credit history, income, and financial dependents. This study contributes to finance by presenting predictive tools that can help financial institutions make more informed and efficient decisions regarding loan approvals. However, we also acknowledge that the Naïve Bayes method has limitations, such as the assumption of feature independence, which must be considered in the interpretation of prediction results.

Optimasi Fitur Seleksi Random Forest Menggunakan GA Dalam Klasifikasi Data Penyakit Gagal Jantung

Diseases of the heart and blood vessels, such as coronary artery disease (heart attack), cerebrovascular disease (stroke), heart failure (HF), and other pathologies, are collectively referred to as cardiovascular disease (CVD). Globally, around 17 million people a year die from cardiovascular disease, with mortality increasing significantly for the first time in 50 years. Has performed an analysis of the performance of the selection algorithm with case studies predicting the determination of the customer's risk profile. Data mining is the extraction of previously unknown or previously hidden patterns from large databases or data warehouses. This study compares data mining classification models Nave Bayes, Decision Tree, Random Forest, KNN, and SVM to find the most effective model for classifying customer profile data. Later, the most accurate model will be proposed as a replacement model for forecasting the customer's risk profile. As a result, the accuracy value obtained is 82.93% and AUC is 0.896. Then accuracy testing is carried out using the rapidminer application. Testing on rapidminer was carried out with the highest accuracy obtained with an accuracy value of 86.64% and an AUC of 0.880.

Managing Peak Load in the Past, Present, and Future: Challenges and Opportunities for a Flexible Energy Future

Taken at face value, peak load management involves controlling or influencing the time of day when electricity is used in homes, businesses, and public facilities. The desire to reduce peak electricity use has been based on the availability of generation resources and their associated costs. During periods of high electricity demand, retail energy providers, including distribution utilities, purchase power from generation sources that are less efficient than baseload generation resources such as hydro, nuclear, or coal. Operating the electrical grid has never been simple, but today the balance of supply and demand is getting more complex. On the supply side, the increasing penetration of renewable and distributed energy sources, such as solar and wind power, makes peak load management more complex. These sources are inherently intermittent, meaning that power generation cannot always be scheduled to meet demand. Additionally, electrification of the buildings and transportation sectors is changing the load profiles of customers and the regions of a distribution utility's service area. Yet, the rise of distributed energy resources (DERs), such as rooftop solar and battery storage, has created both opportunities and challenges for grid operators. Peak load management is rapidly evolving from past practices, with new use cases, economic drivers, hardware and software, and information technologies playing an increasingly important role. A cleaner energy future, while critical to our society and the environment, is not without increasing cost pressures, economic risks, and reliability challenges from extreme weather events. While we know where we are going, it is not exactly clear how we will get there. Nonetheless, the emphasis on peak load management will only increase in scale and sophistication. To better predict and prepare for the rapidly changing energy landscape, this editorial discusses the past and present state of peak load management and how it might be evolving into more flexible load management.

Determinants of Online Food Delivery(OFD) Sales during COVID-19

Purpose: The outbreak of COVID-19 in early 2020 compelled consumers to abstain from dining out and instead use online food delivery (OFD) services. This study aims to examine the determinants of OFD sales in a restaurant during the COVID-19 crisis.
 Design/methodology/approach: We analyzed 139,812 restaurant-level credit card OFD transaction data from January 2019 to June 2020 in Seoul, South Korea.
 Findings: During COVID-19, many restaurants participated and experienced sales growth through the OFD platforms. On the demand side, the composition of customers using OFD services has changed, replacing the demand for dining out. Our estimation results show that the food category, customer composition ratio, and past performance significantly affected restaurants’ OFD revenue during the pandemic.
 Research limitations/implications: The results suggest implications for the restaurant industry responding to changing customer needs on OFD platforms. Restaurants with high-performing OFDs before the pandemic experienced higher sales during the pandemic period. The results imply that experiences to fit customers’ needs toward OFD are likely to help restaurants overcome the losses incurred due to the COVID-19 outbreak.
 Originality/value: This study empirically revealed the effect of store category, past performance, demographic, and geographical customer profile on OFD demand for external shocks such as COVID-19.

Factors affecting omnichannel buying online and return in store: evidence from fast-fashion retail

PurposeThe authors analysed the operations of two synchronised channels by focusing on “buy online and return in store” (BORS) strategies in fast-fashion retail by investigating internal and external factors affecting this omnichannel strategy.Design/methodology/approachThe authors apply a combination of techniques to identify the BORS factors. Firstly, a strengths, weaknesses, opportunities and threats (SWOT) analysis was used to define the operational factors of BORS adoption. The authors then apply analytic hierarchy process (AHP) to evaluate the factors under four SWOT categories for kids, male and female consumer groups. The factors of BORS were then ranked using the fuzzy VlseKriterijumska Optimizacija I Kompromisno Resenje (fuzzy VIKOR) approach.FindingsCombining the SWOT, AHP and fuzzy VIKOR techniques, the authors identified 21 factors in this study. The opportunity that BORS provides for trying in the fitting room for a better convenient shopping experience was ranked as the most important factor, followed by the opportunity to create a loyal customer profile with an easy and well-organised return process. Furthermore, the results reveal that the child consumer group is the most critical of the stated operations factors, followed by male and female consumers.Practical implicationsThe authors described the operational factors and supported the decision-making system of BORS for each consumer group with a priority ranking to realise effective managerial management for fast-fashion retailers and practitioners.Originality/valueThe study contributes to the growing literature on the BORS omnichannel strategy, specifically for fast-fashion retail based on consumer needs.

Modelling and Validation of Typical PV Mini-Grids in Kenya: Experience from RESILIENT Project

PV-based mini-grids are identified as a feasible and, often, only economically viable option for the electrification of Kenyan remote and scattered rural areas, where connection to the national grid is challenging, and the related costs are high, if not prohibitive. This paper presents the analysis of typical Kenyan PV mini-grids by using some results of the work in the project “Reliable, Efficient and Sustainable Mini-Grids for Rural Infrastructure Development in Kenya (RESILIENT)”. After presenting average annual and seasonal daily load profiles of residential and small commercial mini-grid customers identified from the measured demands, the paper introduces the main mini-grid components and their models, including a simplified, but reasonably accurate, model of a mini-grid battery storage system based on the manufacturer’s charge–discharge curves. All mini-grid components are assembled in a scalable and easily reconfigurable simulation model of an actual Kenyan PV mini-grid, and they are implemented for the evaluation of PV mini-grid performance and the potential for expansion and connection of additional residential and small commercial customers. During the validation of the developed simulation model using available measurement data, an empirical approach for adjusting the PV system output power is specified for a more accurate match with the measurements. The presented results indicate the importance of the information on the actual control algorithms and control settings of the mini-grid energy management systems, on the thermal dependencies and characteristics of both PV generation system and battery storage system, and on the availability of on-site measurements of temperature and input solar irradiance. The developed PV mini-grid model can be used for further analyses, such as to study the techno-economic performance of different mini-grid configurations, to identify the optimal sizing of mini-grid components, and to specify efficient control and operation schemes based on the locally available resources.