Case study: IoT lighting system cuts energy costs, improves productivity
As Atlas Global Solutions found out after conducting an energy audit of a relatively new manufacturing facility, the light fixtures were costing them more than aging manufacturing equipment—the focus of the energy audit.
+ Also on Network World: How the IoT keeps Ben & Jerry’s ice cream safe +
Atlas, a global protective packaging company based in Sutton, Mass., knew its 200,000-sq.-ft. manufacturing facility was not as energy efficient as it could be, said Frank Tavares, the global process engineer for the company. He never thought the biggest waste would be from the lights in the building, though.
“That caught everybody by surprise,” he said, “especially with a 3-year-old facility with what we thought was the latest and greatest light fixtures.”
The facility had eight D5 lighting fixtures, which Taveres considered relatively new at the time of the audit in 2012. The fixtures’ energy consumption, however, was 1.3 million KW hours per year.
To reduce energy use, Atlas turned to Digital Lumens and its intelligent LED lighting system. The company upgraded the lighting fixtures in the 3-year-old building near the same time it upgraded the lighting in a 15-year-old facility. For each project, Atlas installed intelligent LED fixtures and Digital Light Agents modular sensors—all connected wirelessly to Digital Lumens’ LightRules lighting and energy management software.
The results were dramatic, Tavares said.
Atlas first upgraded the lighting fixtures in its 15-year-old facility. It is 110,000 square feet and had 105 400-watt metal halide fixtures, which Tavares said is typical of most manufacturing warehouses. The building was using about 160,000 KW hours per year.
Tavares said they knew just switching to LED lights would save energy and money. LED lights are significantly more efficient than traditional lights. According to Digital Lumens, a typical 400-watt metal halide lighting fixture costs about $473 per year, while an LED-retrofit with a Digital Lumens’ CLE fixture costs $154 per year. So, Tavares could potentially reduce his yearly energy expense by $34,000 in the 15-year-old building.
For the 110,000-sq.-ft. warehouse, Atlas installed a combination of DLE and ILE high-bay light fixtures. The estimated energy usage after making the switch was expected to be 18,000KW to 20,000 KW a year. The reduction turned out to be significantly more. The facility now uses about 7,200 KW a year, Tavares said.
Three months later, Atlas upgraded the lighting fixtures in the 200,000-sq.-ft. facility—also with a combination of DLE and ILE high-bay light fixtures. The estimated savings for switching out the D5 fixtures was expected to be about 500,000 KW per year. Since 2013, the building has averaged about 300,000 KW per year—a huge decrease from the 1.3 million KW hours per year it had been using.
“We went beyond what we estimated the savings were going to be,” Tavares said. “The savings are mostly because of the controls [provided in the LightRules software]—the intelligence behind the lighting—being able to control each light fixture pretty much individually if we want or as a group.”
LED lights have transformed the lighting world, said Kaynam Hedayat, vice president of marketing and product management at Digital Lumens. And the LED lights combined with less-expensive and easy-to-deploy wireless networks allowed Digital Lumens to combine the two technologies to create intelligent lights.
“The lights are all LED-based, all have small computers inside them with wireless networks,” Hedayat said. “The lights talk to each other, and through these conversations—through this mesh network—you can interact with the lights for various applications. Basically you can control the lights, and you can collect data from the lights based on the sensors that are in the lights.”
The wireless network at the physical layer is 802.15.4 IEEE, which is ZigBee. Digital Lumens modified it for the scale and resilience required for the applications that run with the light fixtures.
Not only are users better able to control the lights, but now they can devise strategies that enable them to save even more energy, Hedayat said.
For example, sensors in the light fixtures sense occupancy and can be set to turn on only when a person enters the space. Sensors also detect daylight and can be programmed to adjust the brightness of the lights based on how much daylight is in the space.
The sensors transmit the data through the mesh wireless network to the LightRules management software, which can run on an enterprise-based server or hosted in the cloud.
“All of the data comes into our light system, we aggregate it, analyze it and present it to [our customers] in the form of reports, such as energy savings reports and occupancy reports,” Hedayat said.
The software has an open API, so customers can also extract data from the system and import it into building management systems or building automation systems, he said. For example, a customer could use occupancy data in an HVAC system to control temperature. The customer could set controls so that the space isn’t cooled as much during times when people aren’t in it.
Controlling lights based on the time of day
The light sensors and software allow Atlas, whose manufacturing facilities operate 24/7, to control individual lights or a group of lights. The company can control the timing and the dimming based on a certain time of day. For Tavares, that means he can turn off or turn down lights depending on employees shifts.
“We have areas that during certain shifts, depending on what the operation is, we don’t need 100 percent light in the area. So, we are able to program the lights in those areas and bring them down to 20 percent or 50 percent—or completely off,” he said.
Monitoring occupancy
Atlas uses the occupancy sensors to automatically turn lights on and off.
“The lights are smart enough that they know if there’s activity,” Tavares said. “You can program them to turn on if there’s activity in the area.”
The occupancy sensors and the reports generated from the data collected have also improved Atlas’ distribution. Tavares can track how much activity is in each aisle of the warehouse and see where the traffic is—where employees are going to pull product for distribution. Tavares positioned those in-demand products together in the warehouse.
Tracking occupancy “gave me the ability to consolidate product that is picked more often than others,” he said. “And rather than having people going from one side of the warehouse to the other, I can consolidate all of the busy products into a small number of aisles so all of the activity is in the same area.”
Doing that means Tavares lights only a section of the warehouse instead of multiple sections, further helping to reduce energy costs. Plus, it improves employee productivity and helps get the product out faster because workers don’t have to go all over the warehouse to get it.
With the lighting systems in all of the facilities, Atlas has saved 75 percent in lighting energy and gained 20 percent in productivity.
Monitoring equipment usage
Tavares also uses the data collected from the sensors to track equipment use.
“We have meters on each piece of equipment, and we track each one to see how often it is physically running. It allows us to see if we need a new piece of equipment or not,” he said.
That type of tracking prevented Tavares from having to buy a $500,000 piece of equipment. The team originally thought, based on employee reports, a new machine was needed to keep up with demand. The monitoring discovered, however, that the machine in the facility wasn’t being used to its full potential. There were spikes in usage followed by long periods of downtime, Tavares said. What they actually needed to do was shift some of the work and do better planning when assigning work to the machine.
“It’s easy for someone to say, ‘We’re busy. We’re full. We need a new piece of equipment to keep up.’ But once you look at the data on what the machine actually runs, you can see what is really happening. You can see opportunities to do a better job planning and make better use of that piece of equipment,” Tavares said.
Monitoring equipment usage also helps Tavares plan maintenance, so he is proactive about it and doesn’t have to deal with the crisis of unexpected machinery failures.
The lighting market started out focusing on energy savings, but it’s evolving quickly towards non-energy applications, Hedayat said. Other non-energy applications Digital Lumens is considering: security, fire and safety, asset tracking, and humidity.
“Some people refer to it as Internet of Things applications, but for us, it’s all about the data we collect on the [light] fixtures, the analytics we do on the data, and the kinds of reports and information we provide to our customers,” he said.