Smart Building Sensor Network Deployment Checklist

At OpenSensors, we have acquired much knowledge in planning, conducting, and managing real-world sensor network deployments.  Our checklist is based on our experience with field deployments. We have found the checklist to help manage the following risks:

  • Site specific risks: gateway/sensor placement for signal quality, ensuring connections with gateway has been established, ensuring gateways connect to cloud
  • Prepare risk – staging to verify design
  • Maintenance risk  – detecting sensors that have failed (e.g. stopped transmitting)
  • Connectivity of network, connectivity from gateway to cloud, deploying – setting up staging, maintenance – keeping the data flowing
  • Scoping process – identify and ensure impact on key KPIs


Site specific checklist

  • Gather CAD drawing or floor plan
  • Plan gateway placement
  • Identify any mounting and placement issues
  • Test building corners for signal quality
  • Confirm gateway locations and pre-installation requirements with onsite contacts

Staging checklist

  • Order sensors and gateways
  • Develop schedule
  • Plan deployment rollout: floor by floor
  • Plan for safe storage
  • Plan and implement staging to pre-configuring gateways and sensors
  • Assemble Project Kit with all hardware (inc adhesives, screws, mounts, ect)

Connectivity

  • Connect gateways and test connectivity using some sensors
  • Test remote access to the gateway
  • Mount gateways into position and document
  • Test end to end network connectivity with a few sensors

Implement

  • Deploy sensors according to the rollout plan
  • Test end to end

Handover

  • Prepare handover documentation with asset register
  • Review visualisations

Case study: Technology Company Managing Desks

A Regional Manager for Office Experience at a fast growing technology company found himself getting increasingly frustrated by the unending demands for more space but equally also by seeing the space half empty most of the time. Mark had the insight to use sensor technology to minimize the currently very expensive real estate costs and looked for data to help manage and plan prime floor space for their rapidly growing company.

He wanted hard data which would help him solve the problem of “how much space do I need?” in a less political and more objective approach. His situation was made more complex by having accrued a number of companies over the years all with their own separate real estate contracts. Both legacy leasing and new real estate plans were based on more guesswork than real data. It was particularly important to take guesswork out of occupancy as these contracts were going to impact them for the next 10 to 15 years.

OpenSensors worked with Mark’s team and Stuart Imhof from Resource At Work to specify and install two types of sensors: desk sensors and meeting room sensors. The desk sensors are passive infrared sensors (PIR) that are triggered by both motion and heat,  providing  real-time 1:1 space utilisation. Initially we did a pilot with 5 devices but currently they have over 200 desk sensors over 6 floors. Our dashboards display usage per desk and an aggregate of the usage over the floor.

Screen Shot 2017-11-22 at 09.38.48

Example: key metrics for desk and meeting room occupancy

The meeting room sensors count the number of people using individual meeting rooms. OpenSensors dashboards display usage per meeting room and also per floor of usage. Now when Mark’s team gets complaints that there aren’t enough meeting rooms, they can dig into the problem and find a workable solution. After installing the meeting room sensors, the team learned that while the meeting rooms were used a lot, it was by individuals. As a direct result of this insight, they’re doing a refit and they’re adding some kind of individual space where people can make phone calls. Another thing that the team learned was that many people would book the meeting rooms but not use them leaving them unoccupied.

Screen Shot 2017-11-22 at 09.37.50

Example: Desk occupancy distribution chart and under- and over utilised facilities

Negotiating with the business units

With desk sensors, he quickly identified that initial desk utilisation was 36%. With these numbers in hand, the team was able to negotiate with the business units and drive up the utilisation. Mark noted that the data from the sensors helped align the business unit leaders’ goals and costs resulting in better space utilisation without additional cost.

Refit plans

The sensors not only enabled better management of space planning, but provided Mark and the facilities team knowledge to develop a needs-based refit plan. Based on meeting room usage, the team was able to add more individual space and free up meeting rooms for large groups.

Sensor deployment process

Initially deployment planning starts with CAD drawings, location data and floor information from the client. From there, we start making decisions about gateway placements and also tying the sensor IDs to the drawings.

Next our installation and maintenance partner, Stuart Imhof from Resource At Work, and OpenSensors, visit the client’s site for a site assessment where we put up the gateways, and test for signal quality on a floor-by-floor basis. During the site assessment, we walk the perimeter of the buildings, ensuring that we have strong signal strength.

Prior to deployment, we do as much prep ahead of time as is possible, so that when we arrive on site, the sensors are turned on, installed and we can quickly identify if the sensor has successfully joined the network. Our aim is to have very little configuration on site, in order to ensure the least disruption to employees.  For more information on our deployment processes, check out our recent blogpost where our Project Manager, Kevin Mugadza, shares his experiences.

Impact

Overall, OpenSensors and Resources at Work have been successful partners for Mark and his team, enabling them to improve and fully understand space planning management. The OpenSensors space occupancy dashboards allow the team to streamline their work and ensure support for all business units.

 

What does project managing sensor deployments involve?

We catch up with Kevin MugadzaProject Manager at OpenSensors

Kevin Mugadza is responsible for a thorough site preparation inspection, off-site staging of the network, deployment, and ongoing maintenance support. Kevin has a decade of experience managing logistics and risks to provide on-time delivery of a fully operational sensor network.

Based on years of experience managing inventory, supply chains and logistics, he sees the following as the top risks for deployment of sensors networks:

  • Site specific risks: gateway/sensor placement for signal quality, ensuring connections with gateway has been established, ensuring gateways connect to cloud
  • Prepare risk – staging to verify design
  • Maintenance risk  – detecting sensors that have failed (e.g. stopped transmitting)
  • Connectivity of network, connectivity from gateway to cloud, deploying – setting up staging, maintenance – keeping the data flowing
  • Scoping process – identify and ensure impact on key KPIs

 

How does a project begin?

blueprint-964630__480

“My involvement begins when the project is signed off and the client has given us the CAD drawings, floor by floor so that an initial site survey can be done.

At the beginning of a project the main thing we look for is mitigating risks and anticipating potential problems so a project can run smoothly. We start with a detailed site survey. The survey let’s us know where best to position the sensors and gateways for maximum signal strength to be recorded.

The next stage is to make sure the client is ready to receive the information when it’s set up. Prepared with a sample kit, we double check it’s powering up and that all sensors are communicating with the gateway.”

What determines where you would put the kit?

“There are a lot of different variables influencing signal strength depending on where you put the kit and these need to be identified so that the data can be gathered securely and precisely. Successful communication between the gateway and the sensors is based on signal strength and ‘signal to noise ratio’ within its LoRa range. We start by centrally positioning a gateway, and then we place a sensor in each of the four farthest corners of the room. From there, you can tell where the maximum strength of signal lies and where it may register as dropping off.”

What affects a signal?

urban-2589645__480

“It depends on the makeup of the building. Materials, such as metal in the foundations of the buildings, can be one way of influencing the sensor. One of the reasons we use LoRa technology is because it reflects off other buildings so we don’t need a clear path for a good signal.”

OpenSensors does installation in stages, why is that?

“The phased approach is to mitigate against risk. We start off with a small pilot for an early win and to adequately prepare for larger subsequent deployments within the company. Rather than trying to tackle too many things at the same time, having a phased approach makes it easier not only to have a reliable understanding of what you can expect over a period of time, but also how best to manage it. We want to understand what variables there may be in the life cycle of a sensor and whether these differ between each site or over different floors. If a sensor goes down, it needs to be in touch with the structure already in place so that it can be quickly picked up and then dealt with.“

How long does a pilot run for?

“The pilot phase will run from 1 to 3 months. Deployment time is kept to a minimum because we have project kits with pre-configured sensors and gateways ready to go straight from the box. This saves a lot of time and money since we are able to operate onsite with only a small focused team.”

How long does it take to install a sensor?

stopwatch-2648253__480

“This depends on the type but for desk sensors it only takes about 20 seconds per unit. They can easily be attached under a desk discreetly and are unobtrusive enough not to be noticed. They can just as easily be unattached and moved as and when needed.”

How do you support the existing structure within the company?

“We try to be as frictionless as possible throughout the whole process. That’s why our systems are designed to work independently of a client’s networks so we can do a lot of our support work remotely when we are troubleshooting. It’s important to establish a clearly defined route in order for this to happen. Tickets can be raised by clients and we have a dedicated team to deal quickly with any issues that might arise.

You need a reliable set of data regularly coming in, so keeping sensors online and talking to the gateway is essential.”

What sensors are clients interested in using?

office-1516329__480

“The most popular sensors are occupancy monitoring sensors which look at the use of space whether that’s meeting rooms, desks, break-out spaces such as canteens, kitchens, etc. We also use environmental sensors which measure factors such as heat, light, noise, air quality, etc.  We help clients look at the trends within their spaces; how busy that space is, where it is and at what times of the day, days of the week or month is it being used most or least. We also look at what affects where people decide to locate themselves within that space, so if you’ve got a lot of hot desks and find that people are gravitating to particular areas to find a quiet space to work in if they need to concentrate, we will be able to see that.

Desk monitoring comes first and then environmental sensors are phased in after to give you a full overview into what may be affecting people’s comfort and productivity.”

What happens after the initial kit deployment?

“After all sensors and gateways are consistently providing the information needed, we make sure we take time to help the client understand how to use the dashboard and support tools that we have in place to keep their network live so they’re comfortable seeing data coming through and able to interpret it.”  

How do you tell whether a sensor is online or not?heartbeat-2418733__480

 

 

 

 

 

 

“All our sensors are setup to send ‘proof of life’ signals to the gateway at regular intervals wherever they are. This alerts us to any issues quickly and makes sure that everything is as it should be. We help clients manage their assets as comprehensively as they need to

We overlay our visualisations on top of a client’s CAD drawings.

Each sensor can be identified from its unique numbered tag so you can see immediately if a sensor isn’t behaving as it should be. Even if it is in a meeting room, it’s quite easy to narrow down pretty quickly where in the building it is. We get an alert which allows an exact location on the floor plan so that sensors can be managed as efficiently as possible.

Not only does it show us where the sensors are in the building but it also gives us a good overview of their general health. The other thing we can see in the dashboard is battery life.”

How long do sensor batteries last for?

medium-percent-1926832__480  low-1926830__480

 

 

 

 

“There are two different types of sensors; one is plugged directly into the mains whilst the other is battery powered. We check the frequency of messages sent to make sure each sensor is registering at a consistent rate. Most clients require sensors programmed to send a regular ‘heartbeat’ every ten minutes. Based on that, batteries will normally last for 2 years. For greater granularity, it is possible to set the sensor to send data more frequently than every 10 minutes  however, the battery life will then be reduced. For most people, a very comprehensive overview is still possible with a heart beat signal every ten minutes. “

How do you know when a battery needs to be replaced?

“All the sensors have an indicator on the dashboard leading up to low battery life. Every time a sensor sends a ‘heartbeat’, it also sends a battery report making it very easy to see when one needs to be changed well before it stops working. “

When does your involvement with the client end?

“For the life cycle of the project I am there to make sure that our clients find the value in the system installed and that they are happy with the data provided. We are there to help meet their expectations and offer continued support throughout.“

About OpenSensors

OpenSensors has a  multi-disciplinary team to manage the risks for deployment of sensors networks. We have experience with thorough site preparation inspection, off site staging of the network, deployment, and ongoing maintenance support. Our process provides on-time delivery of a fully operational sensor network.

Frequently Asked Questions

question-38629__480.png     question-38629__480.png     question-38629__480.png

What do Space Planners and Facilities Managers really want to know…?

Q: What type of sensors are available?

hook-1727484__480.png Office utilisation monitoring: desks, meeting rooms and breakout areas

hook-1727484__480.png Footfall monitors

hook-1727484__480.png Air quality

hook-1727484__480.png Environmental sensing

hook-1727484__480.png Asset management

hook-1727484__480.png Parking sensors

Opportunities for sensors to provide reliable data inc. averages and peak demand:

  • Desk utilisation
    • How many flexible desks and bookable desks do you need?
    • What are the trends of usage over the course the day, week or month?
    • When is peak usage?
  • Footfall (circulation and breakout areas)
    • What is the usage and traffic for common and breakout areas like toilet facilities and kitchens?
    • What are the trends of usage over the course the day or days of the week or weeks of the month?
  • Meeting room quantity and size
    • What is the right mix of meeting rooms for collaboration and communication?
    • How many and how big should they be?
    • What type of equipment such as projectors, large monitors, and white boards are needed and can be accommodated?
  • Environmental issues (noise, lighting, CO2, humidity and temperature)
    • What impacts employee wellness and productivity?
    • What is appropriate noise level for the tasks employees are performing?

Q: How do sensors detect desk occupancy work?

unnamed

There are different techniques that are used to detect desk occupancy, the most common and inexpensive is based on heat. The sensors look for a temperature reading at 37 degrees because in most offices unless you have an office cat like we do (hello Princess), most of the time there isn’t something warm sitting in a chair unless it’s a person. If there’s body heat registered, the desk is occupied.

Q: Do we have issues around privacy?

icon-2174737__480Well, there are always concerns about privacy. It’s very important for management to be clear with employees why they’re installing the sensors. The sensors are used to understand what the current usage of a space is and how to improve the design, or having already redesigned the space, they’re now trying to determine whether it is getting better results. The people we’re working with are primarily concerned about employee well-being and types of sensors used to  support this are air quality, temperature, noise, light, humidity and CO2.  

We recommend phased implementations vs. big upfront design i.e. encouraging an iterative model but with clear aims.  We don’t believe that ‘smart buildings’ are binary and upfront large investment both with software and especially with IoT projects usually just end up in frustration.  We’ve found that showing information and gathering feedback quickly, really helps inform the design process based around evidence and real data.

Starting from a prioritised list of problems and working backwards to the necessary data and then installing sensors is easier.  We spend a lot of time testing sensors and we are totally neutral in our recommendations. With that in mind, we also encourage a bit of consideration of what the basic infrastructure should be, for instance we use gateways, etc. that work using open protocols to enable future sensors to ‘talk’ to the same network.  

Q: What type of air quality sensors are available?  

Commercially available sensors can measure the level of potential contaminants including; O3, NO2, NO, SO2, CO, PM2.5 and lead.  Most of the devices are easy to connect and provide quality data measurements so that non-technical staff can deploy them.

Q: Is it possible to benchmark for instance comparing occupancy and other metrics between buildings?

images.pngYes. We can tell you, for a set of desks, the average occupancy by the day, week or month. Data is typically sampled every 10 minutes as it’s tagged with the date and time so the aggregate information tells you a lot about the space needs. It’s as if you had high speed, invisible survey takers running around, just making a note of whether the desk is occupied or not every 10 minutes.

Q: How do you select sensors?

We spend a lot of time testing sensors but we are absolutely neutral in our recommendations. Here are some factors to consider in assessing options for sensors.

hook-1727484__480.png cost

hook-1727484__480.png operating lifetime

hook-1727484__480.png accuracy, precision,and bias of measurement

hook-1727484__480.png  range of sensitivity

hook-1727484__480.png speed of response time

hook-1727484__480.png maintenance requirements

hook-1727484__480.png reliability

Q: What is the process of deploying sensors?

OpenSensors recommends a phased approach, from proof of concept to full-scale deployment, to ensure a successful installation of an IoT network in a business environment. Our aim is to reduce the time to go live and minimize risk.

Phase 1 Evaluate sensors:

Evaluate different sensors for quality, signal-to-noise ratio, power consumption and ease of setup by trying them out on a very small scale in a lab.  

Phase 2 Proof of concept:

Do a full end-to-end test to verify that the queries and analytics were feasible by connecting 5 to 10 sensors to a cloud infrastructure.

Phase 3 Pilot phase:

Move out of the lab into your actual environment. Typically, this requires somewhere between 30 to 100 sensors. We suggest a one to three month test to ensure that the sensors work at scale and the gateway can handle the load, similar to production usage.

Phase 4 Plan and implement full-scale deployment:

After the pilot phase, there should be enough data to verify network performance and your choices for sensors and connectivity, after which, full deployment can be planned in detail and implemented.

Contact us if you would like assistance on sensor selection, network design, or planning a proof of concept deployment.