Demand Forecasting Models for Medicines through Wireless Sensor Networks Data and Topic Trend Analysis

Abstract

Demand forecasting in the biomedical area is becoming more important because of radical changes in the macroeconomic environment and consumption trends. Moreover, the need for big data analysis on data from wireless sensor networks and social media is increasing because it shows not only the rapidly changing environmental data such as fine dust concentration but also the responses of potential customers that are expected to affect the demand for a medicine. Therefore, demand forecasting models based on data analysis in wireless sensor networks and topic modeling of buzzwords in blog documents were suggested in this study. First, we analyzed topics of documents from blogs that describe the symptoms of certain diseases related to selected medicines. Thereafter, we extracted topic trends for a selected period and constructed demand forecasting models that consist of topic trends, environmental data from wireless sensor networks, and time-series sales data. The experiment results show that topic trends about medicines significantly affect the performance of demand forecasting for these medicines.

An Estimation Model for Hypertension Drug Demand in Retail Pharmacies with the Aid of Big Data Analytics

“…In order to avoid this barrier we may obtain the monthly scaled word by dividing the number of monthly occurrences of a specific (single) keyword by a proxy for the overall monthly text volume. In case more than one key topics are involved, a topic trend analysis was performed by averaging disease-related topic weights occurring in a social media blog over the same month to obtain a common monthly weighted average [11].…”

“…In terms of forecasting accuracy, there is a limited number of predictive analytics tools, which are based mainly on modelling data accumulated by business data sources. Current standard forecasting methods applied by businesses and discussed broadly in literature seem insufficient to elucidate the impact of external data sources [11]. Another characteristic of data sources is the structure of the data, which comes as structured (fixed format), unstructured (data without a fixed format) and semi-structured (a combination thereof) [1].…”

Coping with demand volatility in retail pharmacies with the aid of big data exploration

“…Considering the rising importance and the vital role that consumer generated data from the internet plays on drug purchasing preferences, Kim et al (2015) underscore the shortage of relevant existing demand planning research as the majority of available studies focus on simple data analysis. For the impact of such complex internet data to be evaluated, Thomassey and Fiordaliso (2006) suggest that data mining and machine learning have been shown to provide better results than statistical models in nonlinear data structures or when complex relationships exist.…”

“…Watson et al (2014) favour the application of advanced VAR models in minimising out of stock rates in pharmacies by emphasising that the assumptions of conventional inventory theories (e.g., EOQ policies) were insufficient and unrealistic. Kim et al (2015) attempted to evaluate the impact of user generated data from social media blogs by employing the VARX model in considering the impact of exogenous variables on upstream demand forecasting on pharmaceutical supply chains. They showed that the strengths of the VARX model reinforced by its ability to simultaneously analyse the impact of all the variables in the system on each other -making it highly adaptable to structural changes.…”

Elasticidades de demanda de un medicamento antidepresivo en Colombia como estrategia para evaluar poder de mercado

“…10 De acuerdo con la tabla 2 se escogió el PIB mensual estimado por el método de Chow y Lin, debido a que los indicadores AIC y BIC tienen el menor valor. Es importante mencionar que esta metodología sugiere el uso de variables que contribuyan a mejorar las estimaciones; en este caso se utilizaron las variables Imaco 11 e índice mensual manufacturero del DANE.…”