Abstract—Upon an active research regarding the influence of environmental factors on the COVID-19 infection rate, that denoted a divided opinion between a positive or negative influence in term of humidity and the SARS COV-2 virus spread, the authors were able to obtain a quantification effect of relative humidity (RH) influence on the spread of SARS COV-2 virus, in Romania, and consequently the influence of RH on the future number of daily COVID-19 positive cases. The discussed quantification was validated by means of computer software, and the aid of a mathematical model that predicts the outcome of COVID-19 disease on a horizon of 28 days, based on a complex neural network structure, trained using consistent data tracking the evolution of daily COVID-19 cases from March 2020 till October 2020, tool designed by the authors themselves. The validated quantification revealed that the maximum influence of RH on the increase of daily COVID-19 infections is between 60% - 70%, due to the temperate climate, while zero influence occurs at a RH higher than 90%. Index Terms—Relative humidity, environmental factors, COVID-19 evolution, prediction model, neural network. I. INTRODUCTION The first case of COVID-19 was reported in December 2019, in the region of Wuhan, China. COVID-19 is a disease due to the coronavirus SARS COV 2, that is a rapidly spreading virus, thus the coronavirus outbreak evolved into a pandemic on 11 March 2020 [1]. The COVID-19 patients suffer from mild to severe respiratory illness [2]. In Romania, the first case of COVID-19 was reported in 26 February 2020, and the national authorities imposed a national lockdown from 16 March to 14 May, that decreased the number of active cases in our country. From 15 May, our country is maintaining a state of alert, established by the President of Romania. However, with the relaxation of restrictions on the territory of our country, a first wave of infections appeared in April, followed in September by a second wave, reaching a much higher infection rate than at the beginning of the pandemic. Since then, our authorities have taken different actions to keep the pandemic under control. The ascending allure on the evolution of COVID-19 daily cases, is depicted in Fig. 1, were it is represented the number of cases reported each day as COVID-19 positive, from 26 February 2020 until 1 March 2021, the web page https://covid19.geo-spatial.org being the source of data [3]. Manuscript received March 20, 2022; revised May 12, 2022. Iulia Clitan, Vlad Muresan, and Mihail Abrudean are with the Automation Department, Technical University of Cuj-Napoca, Romania (e-mail: iulia.inoan@aut.utcluj.ro, vlad.muresan@aut.utcluj.ro, mihai.abrudean@aut.utcluj.ro). Andrei F. Clitan is with the Railways, Roads and Bridges Department, Technical University of Cuj-Napoca, Romania (e-mail: andrei.clitan@cfdp.utcluj.ro). The red dotted vertical lines, from Fig. 1, represent restrictive measures imposed by the authorities, the green dotted vertical lines represent relaxation measures, while the blue dotted vertical lines represent national or religious holidays. ` Since, mainly the lock down measures imposed in order to slow the virus transmission rate, to relieve the pressure on the health care department and consequently decreasing the number of fatalities due to COVID-19, impacted the socio-economic and financial matters of the different countries of the world [4], many research is conducted on trying to understand and predict either the spread of the virus and the categories of vulnerable people, or the prediction of the effect of the virus, and the severity of the illness [5]-[7]. The authors are working on creating a predictive model, which will be incorporated into a web application [8], in order to predict the evolution of the daily number of COVID-19 cases in different scenarios, namely to analyze primarily the influence of measures taken by the authorities, either relaxation or compulsion. It is thus desired to obtain an aiding tool, to facilitate the local authorities in making the right decisions to maintain an equilibrium between controlling the transmission of COVID-19 and the impact on the socio-economic domain. The process of virus transmission is a biological process which has the particularities of having large number of input and output signals. Consequently, the studied process is a strong nonlinear MIMO (Multiple Input Multiple Output) process. For a correct assessment and prediction of the COVID-19 transmission (the increasing or decreasing number of cases), several factors were considered as inputs to the designed mathematical model, for example the number of daily infections, number of daily deaths, number of cures, number of quarantined persons, measures taken by the authorities, including the influence of environmental factors on the evolution of COVID-19 daily cases. Fig. 1. Daily cases of COVID-19 infections in Romania according to geo-spatial.org [3]. Regarding the influence of environmental factors, a more detailed analysis is caried out regarding the relative humidity’s influence on the transmission of COVID-19, and Influence of Humidity on the Evolution of COVID-19 Daily Cases in Romania — Data Based Study Iulia Clitan, Vlad Muresan, Mihail Abrudean, and Andrei F. Clitan International Journal of Modeling and Optimization, Vol. 12, No. 3, August 2022 87 DOI: 10.7763/IJMO.2022.V12.806