Method For Demand Forecasting Research Articles

A Human Resource Demand Forecasting Method Based on Improved BP Algorithm.

Human resources are the first resource for enterprise development, and a reasonable human resource structure will increase the effectiveness of an enterprise's human resource input and output. The reality is that even if an enterprise designs a human resource allocation plan in accordance with the corporate strategy, it is impossible for the enterprise to operate in full accordance with the plan during the operation process, so the human resource allocation plan only reflects the law of the enterprise's human resource needs during the enterprise development process. Giving effective guidance to the specific work of human resources is difficult. It is impossible to carry out effective human resources structure adjustment to adapt to changes in human resources demand due to changes in corporate tactics, business, scale, and other factors, especially when the current domestic human resources market has not yet fully formed. This paper examines the impact of key factors such as the company's business growth scale and production efficiency improvement on human resource needs with the goal of improving team structure, optimizing staff allocation, controlling labor costs, and improving efficiency and benefits. In this paper, we attempt to develop a human resource demand forecasting model based on business development and economic benefits and guided by intensive human resource development. We analyze and forecast the enterprise's total human resource employment, personnel structure, and quality structure using this model. In light of this, this paper employs an improved BP neural network to construct a human resource demand forecasting system, resulting in a new quantitative forecasting method for human resource demand forecasting with strong theoretical significance. Simultaneously, the human resource demand forecasting system developed can enable enterprises to carry out personnel demand forecasting from the actual situation, making forecasting more applicable, flexible, and accurate, allowing enterprises to realize their strategies through reasonable human resource planning.

Tourism demand forecasting with spatiotemporal features

Tourism demand forecasting is a crucial prerequisite for effective and efficient tourism management. This study develops a novel model based on deep learning methods for precise demand forecasting, namely, spatial-temporal fused graph convolutional network (ST-FGCN). ST-FGCN generates forecasts based on spatial effects extracted using graph convolutional network and temporal dependency captured through long short-term memory. A data-driven spatial matrix is used in our model to strengthen forecasting performance further. Two markedly different forecasting experiments verify the effectiveness of our model. Empirical results suggest that incorporating spatial effects can remarkably reduce forecasting errors. Furthermore, our model shows good applicability for data with different time granularity and different periods: before and during the COVID-19 pandemic.

Systematic Review of Deep Learning and Machine Learning for Building Energy

The building energy (BE) management plays an essential role in urban sustainability and smart cities. Recently, the novel data science and data-driven technologies have shown significant progress in analyzing the energy consumption and energy demand datasets for a smarter energy management. The machine learning (ML) and deep learning (DL) methods and applications, in particular, have been promising for the advancement of accurate and high-performance energy models. The present study provides a comprehensive review of ML- and DL-based techniques applied for handling BE systems, and it further evaluates the performance of these techniques. Through a systematic review and a comprehensive taxonomy, the advances of ML and DL-based techniques are carefully investigated, and the promising models are introduced. According to the results obtained for energy demand forecasting, the hybrid and ensemble methods are located in the high-robustness range, SVM-based methods are located in good robustness limitation, ANN-based methods are located in medium-robustness limitation, and linear regression models are located in low-robustness limitations. On the other hand, for energy consumption forecasting, DL-based, hybrid, and ensemble-based models provided the highest robustness score. ANN, SVM, and single ML models provided good and medium robustness, and LR-based models provided a lower robustness score. In addition, for energy load forecasting, LR-based models provided the lower robustness score. The hybrid and ensemble-based models provided a higher robustness score. The DL-based and SVM-based techniques provided a good robustness score, and ANN-based techniques provided a medium robustness score.

Review and analysis of artificial intelligence methods for demand forecasting in supply chain management

The proper selection of a demand forecasting method is directly linked to the success of supply chain management (SCM). However, today’s manufacturing companies are confronted with uncertain and dynamic markets. Consequently, classical statistical methods are not always appropriate for accurate and reliable forecasting. Algorithms of Artificial intelligence (AI) are currently used to improve statistical methods. Existing literature only gives a very general overview of the AI methods used in combination with demand forecasting. This paper provides an analysis of the AI methods published in the last five years (2017-2021). Furthermore, a classification is presented by clustering the AI methods in order to define the trend of the methods applied. Finally, a classification of the different AI methods according to the dimensionality of data, volume of data, and time horizon of the forecast is presented. The goal is to support the selection of the appropriate AI method to optimize demand forecasting.

Supply Chain Inventory Model with Markov Chain Demand

This study aims to develop a new one-vendor multiple-buyers integrated inventory model. We believe that our proposed model can forecast the demand of all buyers in the coming future by using data that have already existed and to minimize the total cost-for both buyers and vendors. In recent days, the Markov chain approach has played one of the critical methods of demand forecasting in the Supply Chain Management (SCM) field. The proposed model of this study is to analysis the demand of all buyers in one specific season that was impacted by the demand in the same season from last year. Finally, the results of this article discover the most optimal number of buyers and shipments, and the quantity of demand per period. Furthermore, a sensitivity analysis is also conducted to find out the sensitivity of the new model.

Forecasting the Cargo Throughput for Small and Medium-sized Ports: Multi-stage Approach with Reference to the Multi-port System

Purpose: The paper aims to elaborate on long-term throughput forecasts in Small and Medium-Sized ports (SMPs). The research problem relates to determining a method that is relevant for the long-term transhipment forecasting in SMPs. Design/Methodology/Approach: Research was applied to the Polish ports system consisting of three major ports (Gdansk, Gdynia, and Świnoujście) and the minor port of Szczecin. Forecasts for the cargo groups in the major ports were produced using regression models where the model parameters were estimated with the Ordinary Least Squares Method. The obtained forecasts of the throughput of cargo groups in major Polish seaports were converted into dynamics indices. The resulting matrices of the cargo throughput dynamics indices established for the leading Polish seaports were used to prepare cargo throughput forecasts for the port of Szczecin. Findings: Elaborated throughput forecasts indicate that, in the future, the port in Szczecin will retain its universal character, and a moderate increase in the cargo volumes confirms that it will serve as a complementary port to the major ports in the range. The forecasted modal split of hinterland transport indicates the increasing importance of road transport, and a decreased rail and inland waterway transport to/from the hinterland. That is likely due to a lower cargo throughput and difficulties in organizing the rail and inland waterway corridors based on smaller freight volumes. Practical implications: The study provides a practical tool for long-term throughput and hinterland traffic forecasting in SMPs. It is addressed to port authorities as the method may be used in decision-making on capacity expansion and academics exploring the phenomenon of predictions in maritime transport. Originality/Values: The novel method of demand forecasting in SMPs includes relations between major and minor ports in the range and multi-staged validation of results. Research advances studies on the dynamics of multi-port systems.

A Comprehensive Study on Demand Forecasting Methods and Algorithms for Retail Industries

Without a doubt, demand forecasting is an essential part of a company’s supply chain. It predicts future demand and specifies the level of supply-side readiness needed to satisfy the demand. It is imperative that if a company’s forecasting isn’t reasonably reliable, the entire supply chain suffers. Over or under forecasted demand would have a debilitating impact on the operation of the supply chain, along with planning and logistics. Having acknowledged the importance of demand forecasting, one must look into the techniques and algorithms commonly employed to predict demand. Data mining, statistical modeling, and machine learning approaches are used to extract insights from existing datasets and are used to anticipate unobserved or unknown occurrences in statistical forecasting. In this paper, the performance comparison of various forecasting techniques, time series, regression, and machine learning approaches are discussed, and the suitability of algorithms for different data patterns is examined.

An Ensemble-Learning-Based Method for Short-Term Water Demand Forecasting

Short-term water demand forecasting has always been a hot research topic in the field of water distribution systems, and many researchers have developed a wide variety of methods based on different prediction periodicities. However, few studies have paid attention to using ensemble learning methods for short-term water demand forecasting. In this study, an ensemble-learning-based method was developed to forecast short-term water demand. The proposed method consists of two models: an equal-dimension and new-information model and an ensemble learning model. The purpose of the equal-dimension and new-information model is to update data for modelling periodically, while the ensemble learning model is used for water demand forecasting. To evaluate the forecasting performance, the proposed method was applied to a data set obtained from a real-world water distribution system and compared with the single back-propagation neural network (BPNN) model and the seasonal autoregressive integrated moving average (SARIMA) model. The results show that the proposed method improves both the accuracy and stability of water demand prediction. The proposed method has the potential to provide a promising alternative for short-term water demand forecasting.

Demand-predictive storage assignment mechanism for flower auction centers

As the number of daily transactions continues to increase, congestion frequently occurs in flower auction centers. Put system is widely applied in intralogistics operations, which includes distribution and redistribution areas. The uncertain arrivals of demands pose significant challenges for the efficient intralogistics operations in flower auction center. In order to improve performance of the put system, this study newly designs a demand-predictive storage assignment (DSA) mechanism in which uncertain demands are forecasted by constructing ratio time series of each customer. Based on the demand forecasts, the customer locations within the distribution area and the number of locations within the redistribution area are easily determined. Furthermore, a paired redistribution strategy is proposed that enables two customers to share a staging block. A simulation experiment bed is constructed based on a real-life case. The experimental results indicate that the forecasting method outperforms other demand forecast methods in literature with lower forecasting error, and the proposed DSA mechanism reduces the total travel distance compared with the closest open location.

Implementation of inventory management in footwear industry

Purpose: The objective of this research was to implement new inventory management in a footwear company through the analysis of indicators obtained from inventory data collection.Design/methodology/approach: The methods of ABC analysis, demand forecasting, safety stock, reorder point and economic order quantity were applied. The items in inventory were classified by order of financial importance through ABC analysis, and the proposed indicators were analyzed to determine the moment the inventory replenishment should be carried out as well as the purchase lot size for each item. The research also analyzed the behavior of the demand and pointed out the demand forecasting method that came closest to reality.Findings: The study presents a method of implementing inventory management based on indicators derived from the application of ABC curve methods, demand forecasting, safety stock, re-fulfillment point, and economic purchased lot. It also indicates how the ABC classification of stocks can be used to check the most representative materials in stock. The study also highlights that the rejection of modifications can be surpassed by obtaining favorable results.Research limitations/implications: The inventory management applied in this work is based on indicators that resulted in two main data which were able to define the size of the purchase lot to be ordered and the amount of material needed.Practical implications: The methods of ABC analysis, demand forecasting, safety stock, reorder point and economic order quantity were applied. The items in inventory were classified by order of financial importance through ABC analysis, and the proposed indicators were analyzed to determine the moment the inventory replenishment should be carried out as well as the purchase lot size for each item. The research also analyzed the behavior of the demand and pointed out the demand forecasting method that came closest to reality.Originality/value: In this study, a method applied is presented, highlighting the importance of the methodological application for the implementation of inventory management. The study contributes to the encouragement and adoption of methodologies to improve analysis and inventory management in companies.

Comparative Analysis of Two Methods of Natural Gas Demand Forecasting

Based on the monthly dataset of natural gas demand and meteorological factors in heating season (November to next March) of Beijing from 2002 to 2013, an analysis of meteorological condition of natural gas fluctuation is done, which used by Empirical Mode Decomposition (EMD) and statistical correlation methods. A novel intelligent prediction model (EMD_BP) of natural gas demand in heating season via EMD and Back Propagation (BP) neural network algorithm is proposed. In addition, we adopt a prediction model only using BP neural network algorithm compare to the above combine method. The results are as follows, (1) EMD method is a better approach to decompose the social demand and meteorological demand of natural gas. (2) There is a close affinity between natural gas demand and meteorological factors in heating season. The meteorological demand of natural gas is significantly negatively correlated with the mean temperature (also minimum temperature, maximum temperature and negative accumulated temperature), but significantly positively correlated with rainfall and low temperature (<-8°C) days. (3) Comparing with BP model, EMD_BP model can accurately fit the changing tendency of natural gas time series. The results show that the EMD_BP prediction model has better applicability and extensive popular prospect.

Hierarchical pattern recognition for tourism demand forecasting

This study proposes a hierarchical pattern recognition method for tourism demand forecasting. The hierarchy consists of three tiers: the first tier recognizes the calendar pattern of tourism demand, identifying work days and holidays and integrating “floating holidays.” The second tier recognizes the tourism demand pattern in the data stream for different calendar pattern groups. The third tier generates forecasts of future tourism demand. Evidence from daily tourist visits to three attractions in China shows that the proposed method is effective in forecasting daily tourism demand. Moreover, the treatment of “floating holidays” turns out to be more effective and flexible than the commonly adopted dummy variable approach.

Intelligent modeling of abnormal demand forecasting for medical consumables in smart city

Abstract Accurate demand forecasting is critical and difficult for managers, especially for complex demand patterns. In this paper, we develop methods for demand forecasting of sparse, transient and erratic medical consumables. Firstly, combining statistical learning of historical data with basic linear regression, price discount estimates are proposed. To reduce sparse estimates, the transformation of historical demand data is added to the linear regression model. Secondly, some general methods are proposed to deal with demand patterns that we cannot clearly capture. Thirdly, we propose optimized model specifications to select optimal model and reduce redundant variables to avoid underfitting or overfitting. In the last, some numerical experiments are carried out based on the model we propose and some completive models in the actual demand data set. In this study, we develop the most realistic price discount response function based on the problem background, which can further improve demand forecasting performance. This paper also discusses many interesting findings and conclusions.

Forecasting and optimisation for microgrid in home energy management systems

The wide proliferation of renewable energy and deregulation of power grid systems require small power utilization systems to deploy intelligent methods of adjustment to the user power demand. To accomplish this goal, the smart power demand forecasting and power consumption optimization methods and algorithms need to be developed. For this purpose, small power utilization systems can benefit from the techniques developed for the smart grid in general. The present paper is devoted to the development of a forecasting model based on the Long Short-Term Memory (LSTM) method and an optimization model based on Genetic Algorithm (GA) adopted for the use in home energy management systems (HEMS). The present work describes a smart microgrid architecture with a focus on LSTM and GA. The experiments demonstrate that the developed algorithms generate a stable pattern of daily power demand. The use of the developed algorithms allows automated shifting of power to achieve the lowest price without sacrificing their comfort. The main contributions of the present work are the inclusion of all parts of the smart microgrid architecture (non-invasive load identification, forecasting, optimization, renewable energy sources and storage elements) in the research proposing a fully automated control in HEMS rather than recommendation based only.

Research on the method of electricity demand analysis and forecasting: the case of China

With economic expansion of China having moderated to a "New Normal" phase, concerns on the surplus supply of electricity have accelerated, especially in the northwest and northeast regions. The ongoing power system reform also brings various of uncertainties, posing challenge to power supply and demand balance. Therefore, the accurate estimation of electricity demand is still inevitable and urgent. In this paper, the causality relationship of electricity demand and the selected factors, namely GDP, population, energy structure, industrial structure and urbanization, is examined by using Stationary, Co-integration and Granger Causality Tests. Then the direct and indirect effects of these factors are investigated via a path-coefficient analysis. Finally, in these foundations, an improved Chicken Swarm Optimization based on Stimulated Annealing, namely Stimulated Annealing Chicken Swarm Optimization, is proposed to optimize the weighting factors of three forms of electricity demand models. The Stimulated Annealing Chicken Swarm Optimization not only inherits the advantages of the standard Chicken Swarm Optimization such as uncomplicated principle, handy implementation and robustness to control parameters, but also can avoid premature convergence to improve the ability of finding the best solution. Case study reveals that the Stimulated Annealing Chicken Swarm Optimization has better predictive ability than other benchmark algorithms.

Demand Forecasting of Retail Sales Using Data Analytics and Statistical Programming

Abstract Forecasting the demand of network of retail sales is a rather challenging task, especially nowadays where integration of online and physical store orders creates an abundance of data that has to be efficiently stored, analyzed, understood and finally, become ready to be acted upon in a very short time frame. The challenge becomes even bigger for added-value third party logistics (3PL) operators, since in most cases and demand forecasting aside, they are also responsible for receiving, storing and breaking inbound quantities from suppliers, consolidating and picking retail orders and finally plan and organize shipments on a daily basis. This paper argues that data analytics can play a critical role in contemporary logistics and especially in demand data management and forecasting of retail distribution networks. The main objective of the research presented in this paper is to showcase how data analytics can support the 3PL decision making process on replenishing the network stores, thus improving inventory management in both Distribution Centre (DC) and retail outlet levels and the workload planning of human resources and DC automations. To do so, this paper presents the case of a Greek 3PL provider fulfilling physical store and online orders on behalf of a large sporting goods importer operating a network of 129 stores in five different countries. The authors utilize the power of ‘R’, a statistical programming language, which is well-equipped with a multitude of libraries for this purpose, to compare demand forecasting methods and identify the one producing the smallest forecast error.

Bayesian BILSTM approach for tourism demand forecasting

The tourism sector, with its perishable nature of products, requires precise estimation of demand. To this effect, we propose a deep learning methodology, namely Bayesian Bidirectional Long Short-Term Memory (BBiLSTM) network. BiLSTM is a deep learning model, and Bayesian optimization is utilized to optimize the hyperparameters of this model. Five experiments using the tourism demand data of Singapore are conducted to ascertain the validity and benchmark the proposed BBiLSTM model. The experimental findings suggest that the BBiLSTM model outperforms other competing models like Long Short-Term Memory (LSTM) network, Support Vector Regression (SVR), Radial Basis Function Neural Network (RBFNN) and Autoregressive Distributed Lag Model (ADLM). The study contributes to tourism literature by proposing a superior deep-learning method for demand forecasting.

Review of Past Urban Transportation Studies and Implication for Improvement on Travel Surveys and Demand Forecasting Methods in Developing Countries

Review of Past Urban Transportation Studies and Implication for Improvement on Travel Surveys and Demand Forecasting Methods in Developing Countries

Power Load Demand Forecasting Model and Method Based on Multi-Energy Coupling

At the present stage, China’s energy development has the following characteristics: continuous development of new energy technology, continuous expansion of comprehensive energy system scale, and wide application of multi-energy coupling technology. Under the new situation, the accurate prediction of power load is the key to alleviate the problem that the planning and dispatching of the current power system is more complex and more demanding than the traditional power system. Therefore, firstly, this paper designs the calculation method of the power load demand of the grid under the multi-energy coupling mode, aiming at the important role of the grid in the power dispatching in the comprehensive energy system. This load calculation method for regional power grid operating load forecasting is proposed for the first time, which takes the total regional load demand and multi-energy coupling into consideration. Then, according to the participants and typical models in the multi-energy coupling mode, the key factors affecting the load in the multi-energy coupling mode are analyzed. At this stage, we fully consider the supply side resources and the demand side resources, innovatively extract the energy system structure characteristics under the condition of multi-energy coupling technology, and design a key factor index system for this mode. Finally, a least squares support vector machine optimized by the minimal redundancy maximal relevance model and the adaptive fireworks algorithm (mRMR-AFWA-LSSVM) is proposed, to carry out load forecasting for multi-energy coupling scenarios. Aiming at the complexity energy system analysis and prediction accuracy improvement of multi-energy coupling scenarios, this method applies minimal redundancy maximal relevance model to the selection of key factors in scenario analysis. It is also the first time that adaptive fireworks algorithm is applied to the optimization of adaptive fireworks algorithm, and the results show that the model optimization effect is good. In the case of A region quarterly load forecasting in southwest China, the average absolute percentage error of a least squares support vector machine optimized by the minimal redundancy maximal relevance model and the adaptive fireworks algorithm (mRMR-AFWA-LSSVM) is 2.08%, which means that this model has a high forecasting accuracy.

Statistical and Machine Learning Methods for Vaccine Demand Forecasting: A Comparative Analysis

This study aimed to find a suitable model for forecasting the appropriate stock of vaccines to avoid shortage and over-supply. The Auto-Regressive Integrated Moving Average (ARIMA) and Multilayer Perceptron Neural Network (MLPNN) models were used for forecasting time series data. The monthly vaccination coverage was used to develop the models from January 2014 until December 2019. The dataset consists of 72 months of observation, the 60 months of data are used for model fitting from January 2014 to December 2019, and the remaining 12 months of data from January 2019 to December 2019 are used to test the accuracy of the forecast. The most suitable forecast model was selected based on the lowest Root Mean Square Error (RMSE) value and the Mean Absolute Error (MAE). The analytical result shows that the MLPNN model outperformed the ARIMA model in forecasting monthly demand for vaccines. The results will help policymakers improve the proper use of vaccination resources.