Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

2023 7th International Conference On Computing, Communication, Control And Automation (ICCUBEA)

Pimpri Chinchwad College of Engineering (PCCOE), Pune, India. Aug 18-19, 2023

Predictive Insight of Global Warming-A


Sustainable Control System
Soumit Roy Hafizullah Mahmudi Rik Das
Analytics Presales and Solution Data Management and Analytics ACM Professional Member
PracticeLead Kamyab ITS Association for Computing Machinery
2023 7th International Conference On Computing, Communication, Control And Automation (ICCUBEA) | 979-8-3503-0426-8/23/$31.00 ©2023 IEEE | DOI: 10.1109/ICCUBEA58933.2023.10392190

Jade Global Kabul, Afghanistan Ranchi, India


Chicago, USA hmahmudi@kamyab.tech rikdas78@gmail.com
Soumit.roy@jadeglobal.com

Jaya H. Dewan Sudeep D. Thepade


Department of Information Technology Computer Engineering Department
Pimpri Chinchwad College of Pimpri Chinchwad College of
Engineering Engineering
Pune, India Pune, India
jaya.h.dewan@gmail.com sudeepthepade@gmail.com

Abstract—Global warming is one of the major issues for Global warming is one of the most critical trending
decades. Implementation of integrated solution to get problems in current times. Identifying the most common
sustainable control system has significant challenges. contributing factors or features that are increasing global
Acceptance of the key precautions and maintenance of warming is one of the key research areas for decades.
sustainability became a bottleneck. The consequence of human
social activities increases the consumption of fossil fuels as The latest models proposed by near past research
power resources which indirectly increased the concentration conclude that earth’s surface warming is accelerated. The
of greenhouse gases and water vapor, in the environment. This current model prediction technique reveals that global
causes the earth's average surface temperature to increase. warming will take a rapid rise. The proposed modeling from
Water vapor is responsible for two-thirds of this issue; previous research advocates 1.5˚C temperature surge is
however, CO2 is proven to be the key factor of global warming. likely to take place by 2030.
The global temperature could have increased to 3.8 ° C in case
CO2 concentration was doubled. Most of the cases researcher were only focused on
predicting Temperature or identifying the co2 emission
The trend and association between rising temperatures and trend but detail solution design on such data engineering is
CO2 emissions are shown in this work, along with an not available in any recent work.
explanation of the many methods behind ideal mitigation. One
of the goals of this research is to identify the issue area and Establishing a solution model for climate change and
integrate the climatic data. This will assist in obtaining a global warming as an urgent matter and striving for
dependable control system capable of addressing global prevention and mitigation by restricting greenhouse gas
warming concerns. This paper outlines how data engineering emissions are key differentiators of this study. In addition to
and analytics may be used to regulate global warming and describing the many principles guiding the best mitigation
provides an architecture for the Green Warming Model strategies, this study will show the trend and association
(GWD). Therefore, the primary contribution of this study between rising temperatures and CO2 emissions.
effort is environment preservation using data engineering and
machine learning (ML). II. RELATED WORK
Business analytics plays major role in Environmental
Keywords— Green House, Global warming, IPA, SARIMAX, management and social science. A pivotal function of green
Data Engineering transformation has been found with business analytics and
I. INTRODUCTION Data science[3]. Making the environment green can done by
applying major steps in environment management.
Indiscriminate human activity is causing serious damage Governance and continuous focus on improvement on every
to our environment. This triggers a set of irreversible step of ecological and manufacturing lifecycle is very
changes to our environment which in turn threatens our important to improve the environmental decay [4]. Making
planet due to social, economic and ecological reason. In the entire supply chain green is the critical step to enable
recent years, we have been witnessing a sudden and environmental management control system[5]. In last 10
dramatic change in our natural environment. Mainly these year researcher have found that analytics and machine
are coming from four reason i.e. Global Warming, Ozone learning is playing key role to streamline and convert the
layer Depletion, Climate change, other environmental supply chain management green[6]. Global warming is one
Degradation [1]. Unless we take some measures to mitigate of the major causes of environmental issue which is invoked
them, these environmental issues will pose a grave threat to by greenhouse gas emission. Earth temperature data and
our feature generation. Of these, global warming is the most greenhouse gas data is in huge volume. Getting 360 degree
important. Earth's temperature is rising due to greenhouse view and actionable insight on high volume environmental
gases trapped in the atmosphere, causing climate change and data can be achieved using big data analytics [7]. Big data
global warming [2]. Building a sustainable environment is analytics can improve the supply chain management to
the key aspect to make the social, economic, and ecological make it green using by optimizing the production lifecycle
indicator in balance. and increasing the sustainability of consumption [8]. There

979-8-3503-0426-8/23/$31.00 ©2023 IEEE 1


Authorized licensed use limited to: PUC MG. Downloaded on May 12,2024 at 17:37:36 UTC from IEEE Xplore. Restrictions apply.
is a deficiency of a system which integrate all aspect of temperature average as a metrics and calculated the trend of
environmental sustainability and provide a 360-degree view it from 1940 to 50.[19][20] found some discrepancy on the
on key measures and provide actionable insights. Most of model for estimation of GMST for last 50 years. [21]has
the researcher has mainly focused on industry and area described internal variation of climate with its internal heat
based approached. This study will mainly focus a transmission methodology to control the production of
sustainable model and a methodology to identify the surplus heat by transiting it to deep layer in ocean.
limitations with cutting edge technologies.
III. SOLUTION TECHNIQUE
Most of the research has worked on the technology
landscape which can be used to get a sustainable Below flow diagram (fig.1) shows high level solution
environment. Inevitably the study of the large-scale, long- methodology for this research work. Temperature data and
term increase of CO2 in the atmosphere, seas, land, and CO2 emission data will be analyzed. Data will be processed
cryosphere is the focus of environmental research. (years, with cleaning techniques to convert this to usable format.
decades, centuries) [9].Warming in Permafrost layer has the Multiple supervised modeling techniques will be used to
prospective to intensify global climate change, because predict temperature and CO2 emission. Key performance
when frozen deposit defrost it unfasten soil organic carbon indicator will be identified to get the key matrices to
[10].It is easy to be skeptical, but pollution is a trending improve global warming. Finally design of one robust
issue and is in attention for all countries. Now it’s the time solution system will be prescribed to control global
to get an integrated and robust solution, these needs to be warming.
acceptable in all administrative level also and it is also
discussed about a low-cost solution using numeric
simulation which can be useful for this problem. But there is
lack of resources on overall solution architecture to get the
complete solution. This study will identify the impacted
region using machine learning techniques and then provide
a 360 degree view to take corrective actions.
The average world temperature is increasing by 1.4 -5.8
degree Celsius. This leads to melting of polar ice caps, rise
in sea levels, shifting weather patterns, increasing flood and
droughts, forest fires, increasing epidemics, food shortage
and other issues [11].Climate change has multiple feature
which are not correlated linearly with temperature
changes,[12]has established a model to find the relation of
climate change with other variables. Change in global mean Fig. 1. Solution Approach
temperature from preindustrial to 2005 is varies
approximately between 0.55° and 0.80°C. Given the variety Greenhouse gas emission is increasing by the year.
in global temperature, proper determination of global Although number of renewable energy sources are in place
temperature variance since pre-industrial utilizing the for use in most of the developed country. It has been
machine learning methodologies with scaling responses can observed that most of the developed country contributes a
improve entire control system and utilization of observation major part of it. USA and European Union con-tribute
and identification processes within the limit of simulations majorly. China comes and just after that followed by Japan
available.[13]mentioned that the differentiation with and Russia. India also one of the top contributors. The
observations is not needed the given sparse observations and below graphs (fig.2) shows the greenhouse gas emission
contrasting use of temperatures (air or sea). Accurately trend by major contributing countries.
determining the overall average temperature of the Earth's
surface in response to changes in radiation levels is crucial
for evaluating the potential danger of climate change caused
by human activities. The use of a statistical method that
employs a random linear responsemodel, based on
observations, is aligned with the long-term patterns seen in
global temperature fluctuations[14]. The research conducted
by [15]examined the relationship between temperature
trends and their associated impacts. The investigation
utilized temperature anomaly data and various temperature
indicators, including sea surface temperature (SST), ice and
glaciers, and sea level measurements, to establish that
warming is undeniably occurring. A study by [16]has
outlined that global temperature is increasing day by day
with the increase of population and other factor for years,
whereas temperature average for last 10 year is not same if
we take average 100 years[17]. Researchers are working on
multiple model on environmental science. The research by
[18] has worked in a multiple phase project to streamline the
model and bring some insights on the trend using machine Fig. 2. Greenhouse Gas emission(greenhouse gas emission 1998-2017)
learning algorithm. They have captured global surface

2
Authorized licensed use limited to: PUC MG. Downloaded on May 12,2024 at 17:37:36 UTC from IEEE Xplore. Restrictions apply.
IV. RESULT AND DISCUSSION
With the increase of population in last few decades
demands of energy got increased. Most of the energy is
produced by natural gas and coal. In recent days coal is
comparatively less used for power generation. The below
graph (fig.3) shows the trend of emission by sources. It has
been observed that Co2 emission by coal has been
decreased. On the other hand emission from natural gas has
increased.

Fig. 5. Mean and Std of emission

Fig. 3. CO2 emission stationary test

Emission data set needs to be plotted at first to check the


category of the data. As this will be used in time series
model seasonality of the data should be checked. After that
statistical method will be applied on the data to determine
the seasonality and trend. If the data set has seasonality then Fig. 6. Time series and forecast of emission
it needs to be converted to stationary data set. Residuals part
needs to be treated separately. From below graph (Fig.4) we After Plotting (Fig.6) MA, ARIMA, SARIMA we found
can see that data set has trend and seasonal effect. In this that SARIMA is providing most promising result. These are
case data set in not stationary. observed that confidence level associated with the forecast
is enough to understand and analyze the time series.
Emission from natural gas in increasing rapidly was
observed from the forecast.
V. CONCLUSION
Global warming is a wide range issue across the nations.
In this study we have shown the result focused on
temperature and emission prediction. After predicting the
emission we have established a solution model for a control
system. This control system has the capability to reduce the
global warming if we can place it and utilize properly.
Governance body has a major role to control this climate
issue. In 21st century with the increase of population
demand for non-renewable energy has also increased so
emission has also been increased eventually. So restriction
on fossils fuel is a major step which has been established
Fig. 4. CO2 emission stationary test through the prediction model.

The standard deviation and mean in below plot (Fig.5) REFERENCES


show that it changes with. So trend is identified in time [1] G. C. Hegerl et al., “Causes of climate change over the historical
series. Also, critical values with 90%, 95% and 99% record,” Environ. Res. Lett., vol. 14, no. 12, p. 123006, Dec. 2019,
confidence levels have been observed for test statistics. Null doi: 10.1088/1748-9326/ab4557.
[2] I. Medhaug, M. B. Stolpe, E. M. Fischer, and R. Knutti,
hypothesis cannot be rejected. So time series is not “Reconciling controversies about the ‘global warming hiatus,’”
stationary. Nature, vol. 545, no. 7652, pp. 41–47, May 2017, doi:
10.1038/nature22315.
[3] G. E. Marjaba, S. E. Chidiac, and A. A. Kubursi, “Sustainability

3
Authorized licensed use limited to: PUC MG. Downloaded on May 12,2024 at 17:37:36 UTC from IEEE Xplore. Restrictions apply.
framework for buildings via data analytics,” Build. Environ., vol. [13] K. Cowtan et al., “Robust comparison of climate models with
172, p. 106730, Apr. 2020, doi: 10.1016/j.buildenv.2020.106730. observations using blended land air and ocean sea surface
[4] H. Zameer, Y. Wang, H. Yasmeen, and S. Mubarak, “Green temperatures,” Geophys. Res. Lett., vol. 42, no. 15, pp. 6526–6534,
innovation as a mediator in the impact of business analytics and Aug. 2015, doi: 10.1002/2015GL064888.
environmental orientation on green competitive advantage,” Manag. [14] E. Myrvoll-Nilsen, S. H. Sørbye, H.-B. Fredriksen, H. Rue, and M.
Decis., vol. 60, no. 2, pp. 488–507, Feb. 2022, doi: 10.1108/MD-01- Rypdal, “Statistical estimation of global surface temperature
2020-0065. response to forcing under the assumption of temporal scaling,”
[5] R. Dubey, A. Gunasekaran, S. J. Childe, S. F. Wamba, and T. Earth Syst. Dyn., vol. 11, no. 2, pp. 329–345, Apr. 2020, doi:
Papadopoulos, “The impact of big data on world-class sustainable 10.5194/esd-11-329-2020.
manufacturing,” Int. J. Adv. Manuf. Technol., vol. 84, no. 1–4, pp. [15] M. Ahmed, “Introduction to Modern Climate Change. Andrew E.
631–645, Apr. 2016, doi: 10.1007/s00170-015-7674-1. Dessler: Cambridge University Press, 2011, 252 pp, ISBN-10:
[6] J. Liu, M. Chen, and H. Liu, “The role of big data analytics in 0521173159,” Sci. Total Environ., vol. 734, p. 139397, Sep. 2020,
enabling green supply chain management: a literature review,” J. doi: 10.1016/j.scitotenv.2020.139397.
Data, Inf. Manag., vol. 2, no. 2, pp. 75–83, Jun. 2020, doi: [16] G. A. Meehl, A. Hu, B. D. Santer, and S.-P. Xie, “Contribution of
10.1007/s42488-019-00020-z. the Interdecadal Pacific Oscillation to twentieth-century global
[7] J. M. Swaminathan, “Big Data Analytics for Rapid, Impactful, surface temperature trends,” Nat. Clim. Chang., vol. 6, no. 11, pp.
Sustained, and Efficient (RISE) Humanitarian Operations,” Prod. 1005–1008, Nov. 2016, doi: 10.1038/nclimate3107.
Oper. Manag., vol. 27, no. 9, pp. 1696–1700, Sep. 2018, doi: [17] A. Dai, J. C. Fyfe, S.-P. Xie, and X. Dai, “Decadal modulation of
10.1111/poms.12840. global surface temperature by internal climate variability,” Nat.
[8] R. Dubey et al., “Examining the role of big data and predictive Clim. Chang., vol. 5, no. 6, pp. 555–559, Jun. 2015, doi:
analytics on collaborative performance in context to sustainable 10.1038/nclimate2605.
consumption and production behaviour,” J. Clean. Prod., vol. 196, [18] K. E. Taylor, R. J. Stouffer, and G. A. Meehl, “An Overview of
pp. 1508–1521, Sep. 2018, doi: 10.1016/j.jclepro.2018.06.097. CMIP5 and the Experiment Design,” Bull. Am. Meteorol. Soc., vol.
[9] J. H. Faghmous and V. Kumar, “A Big Data Guide to Understanding 93, no. 4, pp. 485–498, Apr. 2012, doi: 10.1175/BAMS-D-11-
Climate Change: The Case for Theory-Guided Data Science,” Big 00094.1.
Data, vol. 2, no. 3, pp. 155–163, Sep. 2014, doi: [19] C. K. Folland, O. Boucher, A. Colman, and D. E. Parker, “Causes of
10.1089/big.2014.0026. irregularities in trends of global mean surface temperature since the
[10] B. K. Biskaborn et al., “Permafrost is warming at a global scale,” late 19th century,” Sci. Adv., vol. 4, no. 6, Jun. 2018, doi:
Nat. Commun., vol. 10, no. 1, p. 264, Jan. 2019, doi: 10.1126/sciadv.aao5297.
10.1038/s41467-018-08240-4. [20] K. B. Tokarska et al., “Past warming trend constrains future
[11] K. Albrich, W. Rammer, and R. Seidl, “Climate change causes warming in CMIP6 models,” Sci. Adv., vol. 6, no. 12, Mar. 2020,
critical transitions and irreversible alterations of mountain forests,” doi: 10.1126/sciadv.aaz9549.
Glob. Chang. Biol., vol. 26, no. 7, pp. 4013–4027, Jul. 2020, doi: [21] X.-G. Dai and P. Wang, “Identifying the early 2000s hiatus
10.1111/gcb.15118. associated with internal climate variability,” Sci. Rep., vol. 8, no. 1,
[12] S. Sippel, N. Meinshausen, E. M. Fischer, E. Székely, and R. Knutti, p. 13602, Sep. 2018, doi: 10.1038/s41598-018-31862-z.
“Climate change now detectable from any single day of weather at
global scale,” Nat. Clim. Chang., vol. 10, no. 1, pp. 35–41, Jan.
2020, doi: 10.1038/s41558-019-0666-7.

4
Authorized licensed use limited to: PUC MG. Downloaded on May 12,2024 at 17:37:36 UTC from IEEE Xplore. Restrictions apply.

You might also like