With over 395 million workers worldwide sustaining non-fatal injuries yearly, companies are increasingly recognizing the need to adopt integrated, proactive safety measures. They seek to provide safer work environments overall, reduce risk of accidents and injuries across the site, and improve employee well-being. Traditional siloed approaches have led to key insights being missed due to limited visibility of concerns across the board, reducing a company’s ability to root cause incidents, plan safety measures, and understand improvements. Stricter environmental and safety regulations being enacted by governments, has also led to an increased adoption of Environment, Health and Safety (EHS) solutions across companies’ operations.
In this post, we explore trends transforming workplace safety into more integrated strategies, highlight specific challenges in industries, and discuss how Amazon is building safety into its core with a unified approach to workforce safety starting with Prevention through Design.
Safety trends and challenges
One of the critical challenges facing companies is the sheer diversity of safety risks across different work environments, and even within different sites and functions. For example, construction sites can face hazards such as falls from heights and equipment related injuries. Manufacturing facilities need to contend with chemical exposures and machinery related accidents. In-person inspection at high-risk power plants can prove dangerous. Breach detection and non-compliance with standard operating procedures can risk employee safety across industries. These have traditionally only been addressed by deploying narrow solutions which address specific concerns, but do not enable insight from one area of risk to be brought in to understand others.
On the one hand, technologies such as the Internet of Things (IoT), Computer Vision (CV), Machine Learning (ML), Virtual Reality (VR), and Generative AI have opened up many new possibilities for workforce safety, such as continuously monitoring working conditions or equipment, identifying safety incidents or risks as soon as they occur using CV, or training employees using immersive VR simulations. Companies are also no longer solely focused on physical safety, also recognizing the importance of employee well-being. Equipping employees with the right tools improves their effectiveness, and fosters a work environment that reduces stress and promotes well-being.
On the other hand, instead of only deploying specific use cases, focusing on doing so against a wider safety blueprint – to integrate and contextualize insight from each of these use cases, into a central safety data model – will optimize how safety data is collected, analyzed and actioned. This unified approach improves decision making, better collaboration and communication across the organization.
We will discuss further in this blog, how Amazon is creating a unified approach to safety by bringing together real-world data and visual navigation overlays to help accelerate how its employees can holistically understand safety concerns at a site.
Example: Engineering and Construction – An industry actively looking at integrated safety approaches is Engineering and Construction. Safety and efficiency are the bedrock of successful projects in this sector. Risks range from high altitude work and heavy machinery operations, to managing multiple subcontractors and ensuring site security. Each phase of construction, from ground breaking to handover, carries its own sets of concerns, also directly impacting timelines and budgets. This is where a unified safety framework and suite of solutions really matters, to address the multi-faceted nature of construction safety. It’s designed not just to respond to incidents but to prevent them, integrating continuous site monitoring, workforce training, and compliance management into a seamless data model.
IoT sensors offer operational visibility into every corner of the site, monitoring for structural hazards, equipment malfunctions, and ensuring only authorized personnel are on site. This level of control and oversight ensures that safety protocols are not just in place but actively enforced. Also, any solution must streamline safety compliance, even more crucial in an industry like construction where regulations are not just stringent but ever-changing. Companies in this sector must ensure projects not only meet, but exceed, safety standards with up-to-date training and automated compliance reporting to provide evidence of the same. Deploying a unified safety blueprint in the industry translates into reduced downtime, enhanced yield rates, and most importantly, a safer, more efficient environment.
Example: Semi-conductor industry – Another industry which has specific implications for workforce safety is the semi-conductor industry, where precision and safety need to go hand in hand. In the clean rooms and fabrication facilities, there is virtually no margin for error. Contamination control is just as critical as protecting the workforce from chemical exposures and equipment hazards. Here again, a unified approach to safety is crucial to meet these industry challenges. An integrated architecture enables customers to continuously monitor the environment, and detect hazards early. Comprehensive employee training in immersive environments can be tailored to the unique protocols of semiconductor manufacturing. This means proactive contamination prevention and quick incident response, ensuring both product integrity and worker safety.
IoT Technology enables continuous monitoring of clean room conditions, ensuring particles, gases, and other contaminants are kept within strict limits. Additionally, personal protective equipment compliance can be managed with computer vision, and we can target safety protocols against chemical spills and machinery accidents – all while maintaining the rigorous standards needed in semiconductor fabrication
Turning to AWS for workforce safety solutions
AWS Workforce Safety solutions help customers reduce workforce accidents by implementing a modular but unified blueprint to workforce safety. They allow companies to gain insights to enable safer working conditions, improve adherence to safety protocols, and better address safety compliance and auditing needs.
These solutions allow customers to build against a scalable architecture across AWS capabilities in IoT, AI/ML, Computer Vision and data modelling, alongside integrated specialist partner solutions to address all aspects of workforce safety. This enables our customers to address specific challenges such as using computer vision to detect non-compliance, or ability to use Generative AI to query multiple standard operating procedures, safety manuals and guides, or detect equipment operating risks, and also to bring this insight into a central safety data model. This supports our customers in creating multiple workflows to mitigate safety risks, setup low-latency monitoring and alerts, and using machine learning-based models to detect potential incidents across their operations
Following are sample use cases which can be deployed and integrated into a unified safety data model:
- Spot safety patterns and trends — Monitor data patterns and trends in a visual manner to identify risks, improve safety protocols, and address ongoing compliance and auditing needs.
- Identify high-risk areas of a site with historical data — Overlay historic safety and incident data into an immersive visual environment for safety managers and site designers to identify high-risk areas and implement site designs which help prevent the risk of incidents.
- Identify and respond to hazardous working conditions — Use wearable devices and safety sensors to alert workers of hazardous working conditions so they can take immediate action.
- Enable centralized reporting — Enable centralized reporting and documentation to improve discoverability and adherence to regulatory requirements.
- Enhance workforce training — Train workforce on the most up-to-date Standard Operating Procedures (SOPs) in no-risk virtual environments, improving on boarding time and worker safety.
- Reduce need for physical site inspections — Maintain real-time operational data from IoT sensors to reduce the need for physical site inspections and speed up time to resolve safety issues.
- Ensure workers are following safety protocols — Ensure adherence to safety protocols through CV-based AI models to detect breaches of code (e.g., employees not wearing personal protective equipment (PPE).
Looking ahead
As you look to enhance workforce safety for your organization, your teams may have hundreds and thousands of pages of risk observations and records. During a virtual walk down, for example, valuable time may be lost searching for and identifying the right source of data. Apart from manuals, training material and other content, sensors gather continuous data on potential incidents (such as temperature, gas leakage, vibrations), and video cameras with AI/ML computer vision models can detect and alert to incidents (such as non-compliance to protocols such as PPE or breach detection). It can take time to understand and respond to various incidents without data-based insights, and without an overall view of all safety concerns, it is hard to define impactful improvements. We recommend creating a single pane of glass to consolidate all relevant data sources and leveraging AI/ML, along with AR/VR, to build your own knowledge base and make it readily accessible to teams within your organization.
Solution guidance to keep your workers safe
AWS has developed solution guidance to help companies follow best practices for workforce safety. This guidance includes reference architectures covering: 1) data ingestion 2) data streaming and processing, and 3) data visualization and notifications:
- Data ingestion: Enables on-site data ingestion from IoT, video, PLCs, and documents while providing edge computing for CV and robotic functions.
- Data streaming & processing: Ingests and processes operations data from industrial equipment, video feeds, and IoT devices at an edge location. It then prepares and streams the data to a centralized data lake.
- Data visualization and notifications: Ingests data from AWS IoT SiteWise and a data lake to provide dashboards, 3D visualizations, and risk mapping for monitoring workforce health and safety metrics. It also enables direct data exploration and analysis from curated datasets.
AWS Workforce Safety Solution Guidance
Customer case study: Amazon Workforce Health and Safety
Amazon Reliability and Maintenance Engineering (RME) and Workforce Health and Safety (WHS) are implementing a prevention-through-design program by using AWS services (such as AWS IoT TwinMaker) and partner Matterport to identify, and address potential safety hazards at Amazon’s fulfillment sites. By creating digital replicas of real-world working conditions with embedded historical safety data, WHS can pinpoint design flaws that might lead to accidents or injuries. This includes, for example, identifying maintenance access issues for site technicians and proactively addressing them during the design phase of a site. The program allows WHS to ensure the safety and well-being of Amazon employees and pave the way for continuous improvement in the future.
Amazon WHS used Matterport’s technology in the creation of detailed digital twins of the facility. The Matterport partnership with AWS, and integration with AWS services such as AWS IoT TwinMaker creates a friction-less workflow that allowed the customer to “bind” operational data with their Matterport 3D models to create dynamic, connected, and up-to-date digital twins.
“Implementing prevention through design by creating digital replicas of working spaces at Amazon sites will enhance our ability to identify and address potential safety and maintenance access issues. By simulating real-world conditions and incorporating historical safety data, we can now detect and solve design flaws that previously impeded maintenance activities. Hence, prioritizing technician safety and providing opportunities for review and improvements. Additionally, this technology has proven invaluable in identifying hazard-prone areas due to design deficiencies, significantly improving overall safety. The capability to conduct remote risk assessments and manage operations online has been both cost-effective and efficient” – Shreya Hegde, Senior Product Manager – Tech, Amazon Reliability and Maintenance Engineering (RME)
Digital Twin of site with Matterport and AWS IoT TwinMaker
Digital Twin of site with Matterport and AWS IoT TwinMaker
The images above display snapshots within an interactive Matterport 3D digital twin. When integrated with AWS IoT TwinMaker, these Matterport models can be linked to sensor and other data, creating a powerful platform for facility monitoring and workforce health & safety.
About Matterport
Matterport’s spatial data platform, integrated with AWS IoT TwinMaker, can significantly support companies in improving health and safety protocols. It achieves this by offering intricate 3D digital twins that incorporate real-time data from IoT sensors. This combination allows for continuous remote monitoring and thorough visual inspections, minimizing the need for physical presence in hazardous environments. This integrated solution aids proactive hazard management by providing real-time alerts and insights into environmental conditions. It also enables immersive virtual training programs, preparing workers for emergency situations and orienting them to the space without exposing them to actual risks. In the event of an incident, the platform facilitates comprehensive investigation and analysis through detailed visual and data documentation. Additionally, it streamlines regulatory compliance and reporting, ensuring adherence to health and safety standards, and enhances collaboration among safety teams, management, and external auditors for better decision-making and communication.
Conclusion
In this article, we shared the importance of a unified approach across multiple safety use cases, and the key technologies that are part of integrated workforce safety transformation. We also discussed how Amazon RME and WHS is solving use cases such as prevention through design using a digital twin approach. The immersive visualization of the digital twin can improve communication and knowledge transfer within your operations teams by leveraging “see what I see”. This also allows your teams to optimize the process of identifying and resolving issues more effectively.
For more information on the services and partnerships for digital twins you can Access the Workforce Safety Guidance, Read about AWS IoT TwinMaker and AWS IoT SiteWise and dive deeper into the AWS and Matterport partnership, or contact Matterport to inquire directly about their solutions.
About the Authors
Yibo Liang is an Industry Specialist Solutions Architect supporting Engineering, Construction and Real Estate industry on AWS. Yibo has a keen interest in IoT, data analytics, and Digital Twins.
Pallavi Chari leads Go-To-Market for Industrial IoT Applications and Digital Twins at AWS. She has more than 18 years of experience in product strategy and propositions working with several of the world’s leading technology companies. She has supported industrial customers and partners in transformation efforts working across IoT, Edge Computing, 5G Connectivity, and AI/ML to create business value. She is an Economics graduate with a Master’s degree in Business Administration.
Jon Olmstead partners closely with builder teams at Amazon and supports them as key customers of AWS, including those focused on workplace health and safety. Before joining AWS, Jon served as a digital marketing executive for Caesars Entertainment and he used to lead several web development teams at Zappos. He currently lives on Bainbridge Island, WA, enjoys spending quality time with family, and loves to take in long trail runs on the island.
Shreya Hegde is a Sr. Product Manager Tech with Amazon RME, where she collaborates with AWS, Work Health Safety and the People Experience Teams. Starting her career as a software developer, her passion towards building scalable enterprise products led to a transition into product management. Over the past 17 years, Shreya has launched technical products for airlines industry, healthcare startups, revenue cycle management companies, and US government programs. Shreya is a staunch advocate of women in tech, and serves as a circle leader for the Lean In (Non- Profit) Circle in Austin.
Katie Lameti leads the global AWS partnership at Matterport. She works with teams across AWS and other organizations within the Amazon Partner Network to develop Go-To-Market strategies that empower customers to envision, procure, implement, and scale digital twin solutions that drive value for their businesses.
