Triple Pi Sensing to Limit Spread of Infectious Diseases at Workplace

Jānis Grabis

, Rūta-Pirta Dreimane

, Brigita Dejus

, Anatolijs Borodiņecs

and Rolands Zaharovs

Information Technology, Riga Technical University, Zunda krastmala 10, Riga, Latvia

Faculty of Civil Engineering, Riga Technical University, Āzenes 6, Riga, Latvia

DTG, Ltd., Ganību dambis 24A, Riga, Latvia

Keywords: Predictive Sensing, Workplace Safety, Infectious Diseases, Wastewater Analysis, Organizational Data

Integration.

Abstract: Interactions among employees at a workplace facilitate spread of infectious diseases. This paper proposes to

integrate traditional IoT sensor data, wastewater analysis and data from organizational information systems

for timely identification of threats and adjustment of work activities. The overall approach combining

predictive, preventive and prescriptive capabilities is described as well as the overall technical solution is

presented. The proposed approach allows tailoring of work activities depending on macro and micro

monitoring results in a non-intrusive manner.

1 INTRODUCTION

There are many different types of interactions among

employees at a workplace making it one of the most

frequent places for spread of infectious diseases such

as flue or Covid-19 (WHO, 2019; Koh, 2020). The

organizational structure and processes cause

emergence of localized bubbles either promoting or

limiting the spread of diseases (Shaw et al., 2020).

The Covid-19 pandemic has sparked frantic search

for solutions to ensure safe working environment

(Dong and Yao, 2021). Many of the solutions relay

on massive testing, wearable devices and other

intrusive methods or do not provide timely

identification of threats and response (Margherita and

Heikkilä, 2021;Al-Humairi, 2021; Healthline, 2022).

The cost of safety measures is also significant, and

their cost efficient should be optimized (Patrizi,

2021).

This paper proposes to combine various sensing

technologies to achieve timely and non-intrusive

detection of infection threats and to enact suitable

response mechanisms. These sensing technologies

provide predictive, preventive and prescriptive

https://orcid.org/0000-0003-2196-0214

https://orcid.org/0000-0001-8568-0276

https://orcid.org/0000-0002-4842-9944

https://orcid.org/0000-0001-9004-7889

capabilities referred as to Triple Pi. These

technologies are deployed in organizational context,

and organizational structure as well as composition of

project teams are taken into account to assess and to

limited threats.

The objective of the paper is to propose the Triple

Pi approach and to elaborate interactions among the

sensing technologies. The approach is intended for

companies to provide safe working environment and

business continuity during outbreaks on infectious

diseases if remote work is not a suitable option. The

sensing technologies include traditional IoT devices

such as air sensors and cameras as well as wastewater

monitoring. Data from organizational information

system are also solicitated. The paper describes the

overall approach and introduces early results from the

ongoing project on Covid-19 safe working

environment.

The rest of the paper is organized as follows.

Section 2 discusses related work. Section 3

introduced the Triple Pi sensing process. Preliminary

results are reported in Section 5. The technical

solution is presented in Section 4. Section 6

concludes.

Grabis, J., Dreimane, R., Dejus, B., Borodin

ecs, A. and Zaharovs, R.

Triple Pi Sensing to Limit Spread of Infectious Diseases at Workplace.

DOI: 10.5220/0011747400003399

In Proceedings of the 12th International Conference on Sensor Networks (SENSORNETS 2023), pages 87-92

ISBN: 978-989-758-635-4; ISSN: 2184-4380

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

2 RELATED WORK

IoT technologies were among the first to be deployed

to tackle Covid-19 challenges. Gradually they were

merged with data from other sources to form

integrated solutions.

Wearable devices, face recognition and thermal

scanning have been used to identify employees with

Covid-19 symptoms to prevent spreading of the

infection (Otoom et al., 2020; Petrovic and Kocić,

2020; Bashir et al. 2020). An integrated solution

processes real-time information from sensors and

provides compliance measures in dashboard, which

can be used to monitor and assist in Covid-19

standard operating procedure application across

organization (Bashir et al. 2020). Al-Humairi (2021)

proposes an IoT based smart infrastructure

monitoring system for suspected infection cases real-

time identifying and tracking. The system performs

real-time symptom data collection via thermal

scanning algorithm, it includes facial recognition

algorithm and data are analyzed using artificial

intelligence algorithms. Alternatively, monitoring

methods can be used to monitor compliance with

masks usage requirements (Wang et al., 2020;

Meenpal et al., 2020; Shinde et al., 2022) or CO

level

control (Eykelbosh, 2021; Peladarinos et al. 2021).

The integrated solutions also deploy intelligent

algorithms to enact containment measures. These

methods include customer rerouting (Yang et al.,

2022), access control based on number (Andrade et

al., 2022) and air quality prediction (Mumtaz et al.,

2021). Data integration serves as an enabler of

intelligent data analysis (Duda et al., 2021).

The early detection is a paramount to limiting the

spread of infectious diseases. The current methods

often rely on intrusive methods not always suitable in

environments with elevated privacy requirements.

Additionally, few methods fully exploit the

organizational context enabling to improve tailoring

of mitigation measures.

3 TRIPLE PI SENSING

The objectives of the Triple Pi sensing approach are

to identify threats of infectious diseases as quickly as

possible and to enact measures limiting their spread

and impact in an efficient and non-intrusive manner.

Predictive monitoring is performed to provide early

detection (Figure 1). It is based on wastewater

analysis, which is a non-intrusive method (Zhu et al.,

2021) The macro and micro predictive monitoring is

distinguished because the wastewater analysis

relatively expensive and time consuming. The macro

level monitoring is performed at a regional level and

it triggers the micro level monitoring for a company

(building) if an infection is detected. The infection

detection events trigger prevention activities. The

infection detection at the micro level also triggers

adjustments in company’s work organization as

suggested by suitable algorithms. Depending on size

of the country and national policies, the macro level

could be the whole country, state, region or

municipality. For example, the population size in

regions in Latvia varies from approximately 50 000

to 700 000.

Figure 1: Sensing activities in a safe workplace depicted

using BPMN.

Figure 2: Organizational context.

The Triple Pi sensing is deployed in an

organizational context to account for the impact of

organizational structure and processes on spread of

diseases. The organizational context is defined as a

three-layer graph (Figure 2). The organizational layer

associates employees (or persons) with

organizational units and project team. The

organizational units represent permanent

arrangement of employees while the teams are

Macro predictive

monitoring

Infection

detected

Micro predictive

monitoring

Lock

removed

Infection detected

Prevention

Prescriptive

adjustment

SENSORNETS 2023 - 12th International Conference on Sensor Networks

dynamic. The facility layer associates the employees

with a physical workplace or zone. A zone can consist

of multiple sub-zone. The sensing layer defines

sensors available in specific zones.

Figure 3: Integrated risk-based decision-making.

Data provided by various sensors,

organizational information systems and wastewater

analysis are combined to estimate an infectious

diseases risk level, which in turn triggers actuators

(Figure 3). The Time management system manages

data about the organizational units and the zones, the

Project management system manages information

about dynamic project teams and the Enterprise

calendar provides contains scheduled events and their

participants. Jointly these information sources

characterize expected dynamic interactions among

the employees. The intensity and type of interactions

is combined with IoT sensor measurements and

wastewater analysis results to evaluate the risk. The

risk is either predicted or its current value is

evaluated. For example, ventilation can be adjusted

with regards to expected number of the employees in

a room and current air quality measurement or events

can be rescheduled if the wastewater analysis warning

has been triggered.

4 TECHNICAL SOLUTION

The Triple Pi approach is supported by a safe

workplace platform combining interoperable and

reusable services to ensure business continuity and to

reduce risks of Covid-19 spread at organization’s

Figure 4: Components of the technical solution.

Enterprise

calendar

Time

management

system

Project

management

system

Wastewater

analysis

Interactions

IoT sensors

Actuators

Risk

Triple Pi Sensing to Limit Spread of Infectious Diseases at Workplace

Figure 5: Micro level wastewater analysis.

premises (Figure 4). The platform uses data from the

official regulations, Time management system,

Project management system and real-time IoT data

from the Base station installed on premises. The Base

station is used for wirelessly collecting locally

gathered sensor data (air quality, water quality, spatial

positioning). The platform provides notifications and

enquiries for smart devices as well as adjustments in

the Building management system. Internally, the

platform consists of the Data store (integrates and

stores data of external origin), Data interpretation

framework, Risk prevention framework and

Adaptation framework. The Data interpretation

framework contains the Wastewater analysis module

(processes data about virus particles in the

wastewater), Computervision module (ensures

distancing and waring face masks when necessary),

Spatial positioning module (more accurate distancing

measurement, logging social encounter events among

employees), and Analytics module (descriptive and

predictive analysis of air quality). The Risk

prevention framework contains the Risk level

evaluation module, Proactive risk prevention module

(provides guidance for Covid-19 safe corporate event

planning), Reactive risk prevention module (low

latency risk prevention measures based on real-time

data). The adaptation framework contains the

Notification and enquiry module (notifications and

enquires for the personnel) and Adjustment module

(interaction with the Building management system

for reducing Covid-19 related risks). Various Key

Performance Indicators (KPI) are made visible in the

platform’s dashboard.

At the macro level, the wastewater samples are

taken and monitored at wastewater treatment plants

(WWTPs). However, the data that are obtained from

WWTPs give an overall information about situation

in the selected region. At the micro level, the

wastewater samples are taken directly from the

workplace water distribution network what gives

direct information about possible local SARS-CoV-2

virus outbreaks (Figure 5). The wastewater samples

are taken directly from the collector tank using an

automatic 24 h sampler and later concentrated with a

concentration device. The samples are transferred to

the laboratory and used for direct RNA purification

followed by PCR-based analysis. The data received

are analyzed in relation to other sensory

measurements. The main components of the

concentration device are: feed pump, pressure gauge,

ultrafiltration membrane, flow meters with

controllers, concentration tank and permeate tank.

5 PRELIMINARY RESULTS

The platform implementing the Triple Pi approach is

currently under development and preliminary data are

gathered to evaluate feasibility of the approach.

Figure 6. shows macro level monitoring for the whole

Latvia. It highlights that there are several cities with

level of Covid-19 particles in wastewater. That

triggers an alert to the micro level that micro level

monitoring as well as prevention measures should be

put in place.

SENSORNETS 2023 - 12th International Conference on Sensor Networks

The micro level wastewater results data are yet to

be processed while the air quality is continuously

measured (Figure 7). The ventilation flow is

increased to provide even higher level of air quality.

However, there are several spikes in the number of

parts (ppm) in the air what increases the risk of

spread of infectious diseases.

Figure 6: The macro level monitoring in Latvia (https://

bior.lv/lv/par-mums/jaunumi/notekudenu-monitorings-cov

id-19-izplatibas-noteiksanai).

Figure 7: ONSET HOBO MX CO

logger air quality

monitoring data.

Therefore, prescriptive adjustments are invoked.

In this case, rescheduling of planned events is

performed for the particular room (zone). Table 1 and

Table 2 shows original and adjusted schedules April

27th, respectively. The prescriptive rescheduling

model suggests reducing the number of employees

gathering for the meeting. That results in reduction of

the concentration of CO

in air. The intensity of

interactions is also reduced by limiting cross-team

interactions.

Table 1: The original schedule.

Event

Time

Employees

Duration,

min

Customer visi

9:00-10:00 {E1,E3,E4,E5,V1,V2} 60

Team meeting

10:00-11:00 {E3,E4,E5,E6,E8} 60

Team meeting

12:00-12:30 {E4,E5,E6,E8,E9,E13,

E2}

E – employee, V – visitor

Table 2: The adjusted schedule.

Event

Time

Employees

Duration,

min

Customer visi

9:00-10:00 {E1,E3,E4,V1,V2}

Team meeting 10:15-11:00 {E3,E4,E5,E6,E8}

Team meeting 12:00-12:30 {E5,E6,E8,E9,E13, E2}

6 CONCLUSION

The proposed approach and platform advance the

state of the art by integrating IoT sensing

technologies and wastewater analysis to provide

predictive, preventive and prescriptive capabilities

for limiting the spread of infectious diseases in non-

intrusive manner. The adjustment recommendations

are provided in the organizational context taking into

account interactions among employees as determined

from organizational information systems. The current

research focuses on Covid-19 though the model can

be adapted to different infectious diseases. The

technological foundations of the approach have been

established and data analysis will be carried out in on-

going activities of the research project.

ACKNOWLEDGEMENTS

Project “Platform for the Covid-19 safe work

environment” (ID. 1.1.1.1/21/A/011) is founded by

European Regional Development Fund specific

objective 1.1.1 «Improve research and innovation

capacity and the ability of Latvian research

institutions to at-tract external funding, by investing

in human capital and infrastructure». The project is

co-financed by REACT-EU funding for mitigating

the consequences of the pandemic crisis.

REFERENCES

Al-Humairi, S.N. (2021). Design a smart infrastructure

monitoring system: a response in the age of COVID-19

pandemic, Innov. Infrastruct. Solut. 6, pp. 144.

Andrade, S. H. , R. Negrão, M. Fernandes, F. Costa, and M.

C. da Rocha Seruffo (2022). IoT-based indoor access

control and monitoring: a case study from the

containment measures of the COVID-19 pandemic, Int.

J. Sens. Networks, vol. 38, no. 4, pp. 273–281, doi:

10.1504/IJSNET.2022.122579.

Bashir, A., Izhar, U., Jones, C. (2020). IoT based COVID-

19 SOP compliance and monitoring system for

300

600

900

1 200

20/04 22/04 24/04 26/04 28/04

Air quality, CO2

PPM

time (dd/mm)

Triple Pi Sensing to Limit Spread of Infectious Diseases at Workplace

businesses and public offices, Engineering

Proceedings, 2, 1, 1-6.

Dong, Y., Yao. Y.-D. (2021). IoT platform for covid-19

prevention and control: A survey, IEEE Access, vol. 9,

pp. 49929–49941, doi: 10.1109/ACCESS.2021

.3068276.

Duda, O., N. Kunanets, S. Martsenko, V. Nykytyuk, and V.

Pasichnyk (2021). Information technology platform for

the selection and analytical processing of information

on COVID-19, in International Scientific and Technical

Conference on Computer Sciences and Information

Technologies, vol. 2, pp. 231–238, doi:

10.1109/CSIT52700.2021.9648839.

Eykelbosh, A. (2021). Indoor CO

Sensors for COVID-19

Risk Mitigation: Current Guidance and Limitations,

Vancouver, BC: National Collaborating Centre for

Environmental Health.

Healthline (2022). The Simple Science Behind Why Masks

Work, https://www.healthline.com/health-news/the-

simple-science-behind-why-masks-work, last accessed

04/03/2022

Koh, D. (2020). Migrant workers and COVID-19,

Occupational and Environmental Medicine (9), pp.

634-636.

Margherita, A., Heikkilä, M. (2021). Business Continuity

in the COVID-19 Emergency: A Framework of Actions

Undertaken by World-Leading Companies, Business

Horizons, Volume 64, Issue 5, pp. 683-695.

Meenpal, T., Balakrishnan, A., Verma, A.. (2020). Facial

Mask Detection using Semantic Segmentation, ICCCS

Proceedings, pp. 1-5.

Mumtaz, R., Zaudum S.M.H., Shakir, M., Z., Shafi, U. et

al. (2021) Internet of things (Iot) based indoor air

quality sensing and predictive analytic—a covid-19

perspective, Electron., vol. 10, no. 2, pp. 1–26, doi:

10.3390/electronics10020184.

Otoom, M., Otoum, N.; Alzubaidi, M.A.; Etoom, Y.;

Banihani, R. (2020). An IoT-based framework for early

identification and monitoring of COVID-19 cases,

Biomedical Signal Processing and Control, Volume 62.

Patrizi, N., Tsiropoulou, E. E. and S. Papavassiliou, S.

(2021). Health Data Acquisition from Wearable

Devices during a Pandemic: A Techno-Economics

Approach, ICC 2021 - IEEE International Conference

on Communications, pp. 1-6, doi:

10.1109/ICC42927.2021.9500700.

Petrovic, N., Kocić, D. (2020). IoT-based System for

COVID-19 Indoor Safety Monitoring, IcETRAN

Proceedings.

Peladarinos, N., Cheimaras, V., Piromalis, D., Arvanitis,

K.G., Papageorgas, P., Monios, N., Dogas, I.,

Stojmenovic, M. & Tsaramirsis, G. (2021). Early

warning systems for COVID-19 infections based on

low-cost indoor air-quality sensors and LPWANs,

Sensors, vol. 21, no. 18.

Shaw, J. Day, T., Malik, N., Barber, N., Wickenheiser, H.,

Fisman, D.N., Bogoch, I., Brownstein, J.I., Williamson,

T. (2020). Working in a bubble: How can businesses

reopen while limiting the risk of COVID-19

outbreaks?, CMAJ, vol. 192, no. 44, pp. E1362-E1366.

Shinde, R. K , Alam, M. S., Park, S. G., Park, S. M. and

Kim, N. (2022). Intelligent IoT (IIoT) Device to

Identifying Suspected COVID-19 Infections Using

Sensor Fusion Algorithm and Real-Time Mask

Detection Based on the Enhanced MobileNetV2 Model,

Health., vol. 10, no. 3, doi: 10.3390/healthcare1

0030454.

Wang, Z., Wang, G., Huang, B., Xiong, Z., Hong, Q., Wu,

H., Yi, P., Jiang, K., Wang, N., Pei, Y., Chen, H., Miao,

Y., Huang, Z., and Liang, J. (2020). Masked Face

Recognition Dataset and Application, arXiv [preprint],

https://arxiv.org/abs/2003.09093.

WHO (2022). World Health Organization Homepage,

https://www.who.int/publications/i/item/WHO-2019-

nCoV-workplace-actions-policy-brief-2021-1, last

accessed 07/03/2022

Yang, C., W. Wang, F. Li, and D. Yang (2022). An IoT-

Based COVID-19 Prevention and Control System for

Enclosed Spaces, Futur. Internet, vol. 14, no. 2, doi:

10.3390/fi14020040.

Zhu, Y., Oishi, W., Mauro, C., Siao, M., Chen, R., Kitajima,

M., Sano, D. (2021). Early warning of COVID-19 via

wastewater-based epidemiology: potential and

bottlenecks, Sci. Total Environ., vol. 767, doi:

10.1016/j.scitotenv.2021.145124.

SENSORNETS 2023 - 12th International Conference on Sensor Networks