Michael Pytel

Co-founder and Chief Technology Officer of Fulfilld.io

Science-fiction writer William Gibson’s quote, ‘the future is here, it’s just not evenly distributed’ has been repeated so many times it has become tech industry gospel and at BlueCats we are not inclined to argue. What some industries are doing with Real Time Location Systems and its component technologies such as Ultra-Wide Band are the future, and the tech is right on the cusp of going mainstream.  

As we approach this inflection point, we are getting out the crystal ball in an attempt to peer into the future. We are asking industry leaders eleven questions (because ten was just not enough) about where RTLS is now, where it’s going and where it could go next. 

Avid soccer fan and author, Michael Pytel, is the co-founder and Chief Technology Officer of Fulfilld.io, an intelligent cloud-based, machine learning-powered Warehouse Management System with bundled hardware and software to seamlessly orchestrate warehouse operations and workflow.

  1. What are the top five problems that RTLS (Real Time Location Systems) will solve in the next 5-10 years? 

In North America there is a shortage of warehouse and manufacturing plant personnel. RTLS, or indoor positioning, is going to help solve that problem by increasing efficiency in those workplaces. 

Firstly, RTLS solves the problem of travel time and routing inside warehouses and manufacturing facilities.  By allowing algorithms to decide what path to take and what tasks we’re working on based on the worker’s position, we’ll be able to optimize footsteps and travel time. By relying less on human decision making we’ll be able do more with less. 

Number two, indoor positioning solves the problem of resource turnover and onboarding. A brand-new employee walks into a 300,000 square foot warehouse typically spends three to four weeks job shadowing before they understand the warehouse environment.

“…there is a shortage of warehouse and manufacturing plant personnel. RTLS, or indoor positioning, is going to help solve that problem by increasing efficiency in those workplaces”

RTLS understands where the picker is standing, understands the tasks and the correlating XY-coordinates of all the tasks they need to work on. With indoor positioning we can take a day one employee and navigate them through the warehouse. Instead of relying on tribal knowledge and job shadowing to onboard that new resource, RTLS enables new hires to be efficient on Day One rather than Day 30. 

I don’t know if the Robotic Industry Association (RIA) and OSHA have fully ironed out the rules for humans and robots working together but I think the third challenge is handling the coexistence of machines and human beings in warehouse and manufacturing environments.    

There are certain laws and rules about autonomous equipment needing to stop operation when a human being is nearby. Indoor positioning enables tighter integration between robot and human. For example, in addition to its usual sensors an automated forklift can be programmed to stop operation if a human being is working around it.  

Looking outside of warehouse and manufacturing, indoor positioning could change the retail experience by helping people find things more effectively. It could also enhance sporting experiences by helping people navigate large venues and stadiums to find the restroom, restaurant or the activity they’re scheduled for. It sort of becomes the indoor counterpart of Google Maps; how do I get to the restroom instead of how do I get across town?

  1. What is the best use of RTLS to date?

Within warehouse logistics, it’s employee navigation, laying out the best route to optimize a picker’s footsteps. With paper-based pick tickets, the picker grabs a stack, picks those things, comes back to get more pick tickets and so on. There’s a lot of wasted footsteps, a lot of wasted travel time. Optimizing employee travel time is going to optimize the energy the picker expends throughout the day and create a safer work environment when it comes to collision detection with automated equipment. This leads to pickers having a better experience day to day and they leave work feeling they’ve accomplished something.   

We are still in the pilot phase, however, we are working with an automotive manufacturer. Automotive manufacturing plants are very large, 500,000 square feet or more, with workstations for each job whether that’s installing the interior, the engine or exterior components. As there’s not enough space at a workstation to store every part for every car, travelling racking moves components from the warehouse to the manufacturing line. RTLS enables the customer to track the position of those components as they go from warehouse storage to the production line and ensures the racks are at the right workstation at the right time, which greatly reduces line stoppages.  

Another pilot is with a poultry, pork and beef producer. It used to rely on manual checks to make sure the product is refrigerated and frozen, etc. By using indoor positioning, it can measure in real-time how long a product or pallet has been in the freezer. When does that product need to be moved to cold storage or a shipping container? By understanding the XY-coordinates, instead of a human writing it down on a piece of paper and remembering, the company has those products for exact amounts of time. 

  1. What are the top five problems that RTLS will solve in the next 20 to 30 years? 

RTLS is going to be vital for the cohabitation of robots and human beings. Whether it’s in a retail, warehouse or manufacturing space, robots need to understand the position of other robots and human beings within a building.

In 2019 Amazon commented that fully automated warehouses were more than 10 years away. For the 50,000 other warehouses in the United States that are not Amazon, even a bit more than 10 years is too soon for those warehouses to be automated. Robots still don’t have the dexterity of a human hand so while ultra-wideband is accurate enough to get a picker in front of a product, is it good enough for a robot to reach out and grab one bolt? It’s not at the moment and I don’t know if it will be in 10, 15 or 20 years’ time either.  

We have partnered with a company that produces an automated forklift with LiDAR, and computer vision. It can drive, navigate, find an open bin and place a pallet in that bin. But we need to know the indoor position, the XY-coordinate of that pallet so other robots, pickers etc. can find it. Ultra-wideband is 100 percent critical to doing that. 

It is not my industry but contact tracing. By using indoor positioning, we can enact rules that enable organizations to route employees away from each other. If we have another pandemic, rather than organizations or manufacturing floors having to shut down, we could route employees on tasks that keep them physically distanced.

  1. What is driving RTLS adoption in industry right now? 

Optimization and efficiency, moving goods more efficiently through understanding the actual location of those goods. In North America, a lot of overdue products are coming in from Asia and warehouses are at 110 percent, 120 percent capacity. When that happens, people store things in locations they would not normally be stored, and things get lost.

By using indoor positioning and understanding the XY-coordinates of everything warehouse staff have moved and where they have left it, things don’t get lost. Customers want to adopt RTLS to optimize employees and optimize where they are placing products in their warehouse. An employee may see an open space on a shelf, but it may not be the optimal space to store it. Is it a fast moving good we need to keep low and close? Or is it a slow-moving product we can put high and distant?

By using indoor positioning and understanding the XY-coordinates of everything warehouse staff have moved and where they have left it, things don’t get lost.”

By using RTLS, algorithms and machine learning can recommend where to store products because the system understands the physical space. This is something machine learning could not do before because the system didn’t know that this aisle and this bin and this aisle and another bin are 50 feet or 500 feet apart. With indoor positioning, we understand physical space and dimension, we know travel time between two locations. That is enabling a whole new set of algorithms to be written to support decision making in the warehouse.

  1. What will drive RTLS adoption in your industry in 10 years’ time? 

Regulatory requirements. To prevent accidents RIA, OSHA and the world’s other safety organizations will mandate a robot stops work when a human being is nearby. Just like we have airbags and seatbelts in cars, autonomous robots and industrial equipment will have RTLS as a required safety features.

  1. What is holding back RTLS adoption now? 

Lack of education and understanding; it almost sounds too science-fiction-like. Also, a lot of organizations think that RTLS is really expensive [but] in the grand scheme of the systems and the technologies we implement in manufacturing and warehouse environments, indoor positioning is not an expensive proposition. It’s lower cost than many of the applications our customers use within their IT environments, and it provides actionable insights once the customer understands the deployment and the data they’re getting. That’s the value added to the decision making once they understand location.

[There is also a legitimate fear of being tracked]. We’ve built ultra-wideband into an industrial scanner because we felt that people wouldn’t want to wear a lanyard, or a vest being tracked. But by making the ultra-wideband part of the tool that’s helping them do their job, they can put it down and go use the restroom or go to the break room and they’re not being tracked.

[There also needs to be better] messaging around [how] location data can be anonymized. It can be used purely to power algorithms to build better routing for employees in the warehouse and as software makers we can choose to enable those features. It can be used for good and for evil. You could track the efficiency of individual employees and then reward them or incentivize them based on that behavior. Or the data can be anonymized and shared amongst all different types of organizations to optimize routing and planning of employees.  

  1. What is likely to hold back RTLS adoption in the future?   

Confusion as to what indoor positioning technology is best. There are different use cases for everything; you have Bluetooth Low Energy, Ultra-wideband, Google’s Wi-Fi RTT and the ever-increasing accuracy of GNSS. There’s a variety of technologies available and the one thing that could prevent RTLs adoption from growing is confusion in the marketplace as to which technology a customer should use.

  1. Bluetooth is now ubiquitous, UWB is just getting started but how will mainstream adoption of UWB impact your industry, business and society at large? 

It’s a good question and it made me ask myself, ‘have there been other consumer technologies that have weaved their way into business?’ The answer is the smartphone. The Motorola PageWriter and the BlackBerry were business first but then the iPhone was created [and it and other smartphones have spread into business]. I’m ecstatic about the adoption of UWB in smartphones. As the technology is consumerized, the more people understand it and have less fear of it, the more it will solidify its position as the leading technology for indoor positioning. As a startup, our organization Fulfilld.io, is betting on ultra-wideband as the leading indoor positioning technology of the future.

  1. What are the most important issues in the next 10-15 years when it comes to the interaction of humans / people / AI and RTLS systems? 

Understanding indoor positioning opens up the world to have new algorithms created to support better decision making inside warehouses and manufacturing facilities. Machine learning algorithms today don’t understand physical space, the physical dimensions of the warehouse or manufacturing plant. They understand the product and that there are 50 bins available, but don’t know the location of the bins. By using RTLS to understand the XY-coordinates of everything in a facility, these algorithms can be optimized to real world environments. Up until now, this is what’s been missing.

By using RTLS to understand the XY-coordinates of everything in a facility, warehouse management machine learning algorithms can be optimized to real world environments.

  1. In what way does RTLS/UWB represent a ‘killer app’ for you?

We have a lighting manufacturer client with a 300,000 square foot warehouse, 40-foot ceilings, more than a million different light bulbs. A day one worker walking into that warehouse does not know where they are going or what they are doing. The killer app is having RTLS on a device that routes that employee to the first, second, third, fourth, fifth task and have it be context aware. That means if this person is not certified for a forklift, the system won’t give them a forklift task. It will give them hand picking tasks only and route them around the warehouse safely without running into a forklift, robot or any other piece of machinery.

  1. Interoperability will be a vital piece of the puzzle if UWB is to become ubiquitous like Bluetooth. How important are standards to this happening? 

Standards are needed. Today, anybody can build an app that uses Bluetooth accessories and communicate with those accessories using defined protocols. The ultra-wideband chipset is relatively standard, but there are different ways of writing the algorithms to calculate time of flight and position in the warehouse, and those are different by vendor. There needs to be a consortium to develop open-source location algorithms so that software developers can use a single standard of how to calculate the XY-position in a warehouse, a manufacturing facility etc. rather than it being vendor specific. That will be a key component of UWB becoming as ubiquitous as Bluetooth.

Historically, passive RFID is a prime example [of what standards can achieve]. When I started in the industry, there were two big players, Simple Technology and Matrix, that both had their own air interface protocols. Both technologies worked, but there was a reluctance among retailers, who wanted to use that technology, [as they were afraid of lock-in] and [the duopoly] tended to stifle any improvements in the technology. When the industry organization launched a revised air interface protocol independent of the main players, it led to more competition and improvements in the technology, potential customers had choices and that became the impetus for more end users to adopt the technology and it grew.