Faulty data creates monumental decision mistakes

August 31, 2021

Comments read at Churchill Council meeting 9-13-2021

The objective of my statements in this presentation is to lay out why the traffic statistics of the applicant are grossly underestimated.  And since the decisions on related consequences, the noise study, the soot study, the street congestion, etc., are based on these inaccurate trucking numbers, all the reports and studies of   are suspect, and should not be used in any decisions related to this project until the borough is able to do its own independent analysis.

Given the disingenuous data supplied by the applicant, and the high probability for significantly more traffic generation than projected, the applicant has not provided what is required for the approval of this project.

This project is what Jeff Bezos calls a one-way door.  There is no turning back. The consequences are permanent.

So that there is a clear understanding of why the Hillwood traffic counts are not accurate, it is vital to understand how this facility actually operates.  Only then can the conclusions be appreciated.

Documents supporting everything covered, can be found on the Churchill Future Website, including the books, the videos, publication articles, suppliers’ content, association studies, nonprofit studies, and thousands of back links to original sources.  I recommend those with limited time to read the latest Stone book on Amazon itself, and if you are business oriented, the Charan book on the Amazon Management System.

Now, I do have more than casual experience with distribution.  You will want to know this, before I discuss the vehicle counts.  I have worked with truckload and LTL shippers since 1983.   I have imported containers and break-bulk distributed products at the California warehouse since 1998.  I have knowledge in transportation stretching back to the mid 70’s when I worked for Emery Air Freight, so I know a little something about both trucking and air. I can discuss how the given density of an NMFC class will affect rates for any LTL trucking route for a particular product, as well as where to stack an AAY container for use in a narrow-body freighter aircraft, specifically 737s.  I am no expert; I just have some experience.

There has been absolutely no discussions around actual trucking requirements to support this mega distribution center, so to understand the true requirements, it helps to understand the basic operations within such a facility.  This generation of distribution centers is the largest ever constructed anywhere in the world. So, the term Mega is appropriate.

Amazon has different distribution warehouses, each with a different function.  The flow of trucks will be different depending on that function.  You can think of it as a series of cascades.  The Churchill facility will be the last large warehouse in the cascade of products, prepping for the last mile delivery.  Permanent stock will be limited to quick turnover items as determined by Amazons AI algorithms.  At this facility it will mostly be items boxed and labeled for final delivery from previous warehouses.  Thus, it will require more in and out activity than a warehouse whose purpose is mostly stocking products.

An example of the cascade is the Pomegranate Molasses I ordered yesterday afternoon and delivered at 5 today.  It shipped from BDL2, Windsor Ct, to EWR8, Teterboro NJ, to PIT5, Chartiers off I 76, to the USPS who did final delivery to my Penn Hills office.

It is the final breakdown for last mile delivery that influences the large number of trucks needed.

Amazon is unique.  Only other big box stores like Walmart and Target come close to imitating the efficiency of Amazon.  I will simplify the procedures of this distribution center as it is actually quite complex.

This project is Mega because the average footprint in North American for a distribution center is 180,000 sf verses 639,000 sf. with this Hillwood project.   Every additional sf increases exponentially the amount of storage the facility can handle.  Add to that a height around 100 feet verses 40 feet and the mass is easier to visualize.  So, this building is the equivalent of 5 single level warehouses, one on top of the other.

Consider that the traditional distribution center requirement includes storing, picking and packaging.  The amount of merchandise that is the thruput is limited by the number of employees and amount of automation.  The automation at a traditional distribution center is more difficult than simply sorting boxes as at the proposed Churchill facility.

At all stages of the logistics, Amazon has perfected the process of integrating robots for specific tasks, and AI (artificial intelligence for those not sure of the term) for automated decision making.  The goal at Amazon since the beginning is build-to-scale.  The larger the better as fixed costs can be averaged over more data points, continually lowering costs.

One cannot but admire the ingenuity and creativity of Amazon in this aspect.  That is the only way Amazon could come up with the 2-day delivery time constraint, and still keep the costs reasonable. It is amazing.  Even with Prime, Amazon has a 19% gross margin in shipping.  That is incredible.

What we have here in Churchill is not a traditional distribution center.  It is a final sort, before delivery to the final mile customer delivery.  Among all the different fulfillment centers within the Amazon distribution world, this final sort is the most automated of all, for several reasons.

In simple terms, everything coming into this mega warehouse will be shipped out for local delivery in Pittsburgh East either directly from this location, or through several local last mile facilities such as in North Versailles.  Everything in must ship out and do it within 2-day, 1-day, or 2-hour timelines.  And everything shipped out needs new stock shipped in to replace it.

And 1 day is becoming the norm.  Sunglasses I ordered yesterday, arrived this afternoon, from CLE2 to PIT9 to my office.

First, and there are historic ratios used, the number of employees to handle a warehouse of close to 3 million sf would have been about 1200 per shift. Amazon has that down to just 500.   As background the industry ratio is 1 employee for every 2,000 sf.

Now it is also important to consider that there will be in any distribution center, a given number of people who are not, actual employees. For example, security, both at the entry gates and in monitoring security cameras, are almost always contract employees.  Others who maintain equipment are usually employees of those manufacturers or specialist in that sophisticated equipment. Also, most Amazon warehouses hire normal employees through outside companies who do interviewing and training on site and will also need their own parking accessibility.

The IT dept is likely to be Amazon employees here because of the confidential nature of the proprietary programming.  For many facilities, however, the IT is also contracted out.

So where will the majority of these 500 estimated fulltime employees be working.  A large percentage will be controlling the arrival and departure of the freight, the jockeys. Something that still needs humans to do.   The actual yard jockey job description of Amazon will be posted as an exhibit.

The way trucks are managed on site is a whole different world.    Gone are the days when a trucker just showed up and was directed to a loading dock number and backed up in his own rig.  This is long long gone.

Time spent unloading and reloading the actual trailer is separate from how that trailer actually shows up at a given dock.

To give some examples we must first understand the time is money.  2 men may take 2 hours to unload a traditional loose pack trailer.  Adding more men to the job will speed it up, but when you exceed 5, the productivity decreases with so many workers going in and out of the trailer.  They simply get into each other’s way.

To truly appreciate how much stock can be placed inside a standard 53-foot trailer, don’t walk it outside, go inside, all the way to nose.  You WILL be impressed.

Another area that Amazon is perfecting is the ability to quickly load or unload trailers.  It is adapting different methods for different products and other constraints.  Without going too much into detail, one way is to containerize like airplanes, where containers or pallets are pre-loaded elsewhere, and then pushed into the trailer, in minutes, or the reverse when emptying.  Photos of this is another exhibit you can refer to for a visual of the process.

Amazon thinks big picture.  Not just that loading point.  It considers the unloading at destination when configuring the loading method, as well as the next step in the fulfillment sequence. This is sophisticated and complex logistics planning.  One is tempted to think of just how to load and how fast to load a trailer, but that is wrong.  Loading a trailer is just one step in the total logistics system.  Why that is important for you to realize will be evident in a few minutes.

Every activity is related to what took place earlier and what will take place subsequently. This apples to every single moment that both the trailer is handled, and merchandise is handled.

So far, we are talking about a trailer, at the dock, with the door open.  Now, we must think about all the individual things that made that happen.

Let’s first go back to the truck carrying its load from the previous loading location, LEX3 or CLE2, two frequent origins for what I order through Amazon.

At Amazon a driver not from a previous Amazon facility must alert the destination 24 hours in advance of its arrival time at the destination.  In route all kinds of things happen to slow down the movement, from construction delays to mechanical problems to the driver running out of hours.  The AI that Amazon uses takes all these things into consideration by analyzing every route in the US and applying that data for its own estimate of arrival time.  That data allows the receiving facility to estimate with amazing accuracy when that truck will arrive. The driver calling in, for the most part, just confirms what the algorithm predicts and allows minor adjustments.

The question you may have is why?  Every minute of every activity within an Amazon facility is measured.  That includes the loading docks.  Each is kept productive through the AI programming so that there is no down time.  Every trailer is assigned a specific time to be loaded, or a specific time to be unloaded, 24 hours a day.

Semi-trucks arriving at the Churchill facility will almost all be from other Amazon facilities and the time in travel already calculated and conveyed to destination.  this is similar to airplanes with specific arrival times and gates calculated before it the plane even takes off.

Because this facility will be primarily for small package sortable, drivers of the semi-trucks in will be coming from other Amazon distribution centers, no more than 8 hours away.  That is why even the shiny new cabs Amazon leases to those independent drivers have no sleeper as one would expect for long haul trucks.

This is critical because every minute wasted at the loading and unloading of the trailer is costly.   Delays cause ripple effects and additional costs all along the line. For example, a specific item on that truck can be prescheduled to ship 1 hour after arrival.  If it is late, then a commitment to the customer will be late.   Amazon is fanatical about delivery commitments.

We come back to the need for absolute efficiently at the loading dock.  That efficiency is complex and highly controlled.

When every driver arrives at the facility entry gate, at the time estimated, and within the window of his allowed entry time slot, He is then given a location to go and drop his trailer.  Why?  Not being familiar with the facility, he will waste time looking for and backing into the assigned dock, which would have been the norm 10 years ago.  He goes instead to a specific location that is both easy to reach, and easy to drop his trailer.

When you look at a given Amazon facility from the ground or from Google maps, there are hundreds of trailers parked on the perimeter, all the time.  Some empty and some full, some to be unloaded, and some to be shipped out.  That is necessary because the docks operate nonstop, and those trailers are the buffer for that to work efficiently.

This may seem inefficient, to drop his trailer; it is not.

What the semi cab does then we shall discuss later.  Now since every truck entering, is doing the same thing. We have to wonder how a detached trailer gets itself to the dock.

What we have here are the jockeys.  These are drivers, increasingly with specialized towing vehicles, who are experts in re attaching these dropped trailers and putting them into the loading dock position that the AI assigns to that specific trailer.

As a side note, the jockey trucks used are considered off road vehicles, and do not have the same high pollution controls of regular trucks for diesel emissions. And remember, they are operated 24 hours a day. They were not considered in any of the Hillwood studies.  Although electric versions are being developed, they are not yet commercially available.  Exhibit V 3 shows two manufacturers of these trucks, as well as the Amazon job description for the job. Check it out, it is interesting.

And that assignment can vary, just as aircraft may be assigned a gate and then need to come into an alternative, due to some unforeseen situation.  If you fly a lot, you know exactly what I mean.

Let’s take a moment out and think about the how and the why the 49 docks are arranged as they are at the proposed Churchill facility.  Remember, this facility is not unique.  It was specifically designed for final sort.  Meaning, that the primary internal functions are tied directly to the products coming in and going out to minimize handling and for maximum efficiency.

For example, if a specific pallet of products is scheduled to leave out of dock 21, it would not be unloaded at dock 49. But closer to dock 21.  This advance scheduling is incredibly complex with massive data requirements.  The results are beautiful for a logistics-oriented person.   Only Amazon has so far been able to master this process, although reports show that Walmart is close behind.

Now back to the trailer in the parking stall.  That jockey will quick attach and pull that trailer to its assigned location. At that point it must be chocked, an OSHA regulation to prevent it rolling, the doors opened, and the dock leveler lowered.  Because Amazon operates every single minute 24 hours a day, Amazon rarely closes the dock door like most distribution centers.  Although things do happen, and they are fully capable of a total lock down in seconds.

These jockeys are running to and from continually during their shifts.  Time on Task applies here just as it does inside the facility for people boxing items.

Now, let’s talk about the facility itself, and why Amazon is able to operate at such high efficiency with increasingly fewer employees.

This is final sort.  That means the majority of stock is coming in already boxed with labels.  These are small boxes, functioning similar to how UPS originally operated with its belts and paddle arms.  The goal is to quickly assemble shipments into final delivery zones.  A zone is defined as what will fit into a flex driver’s delivery vehicle.

Let’s diverge a moment and think about how that box got to your front door.

First, we talk about the delivery zone. Travel time is lost time. Amazon has broken the entire country   into continually changing zones based on volume of packages that need to be delivered.  This is an incredibly complex algorithm taking into consideration minute by minute traffic flow, population density, weather, even the storage space of the flex drivers vehicle; thousands of variables using data only AWS  (Amazon Web Services, their computer farms) could collect and interpret.  As delivery zones are constantly being updated, it influences the entire distribution system from what is ordered all the way down to the facility planed for Churchill.

Now back to that building.   Most packages are grouped by zone for redistribution to the final mile location.   Understanding how all of this data is interrelated will help explain the traffic volumes.

Ever wonder how that package showed up on your step at midnight?

First, we must understand that Amazon never shuts down, never. There will be inevitable bottle necks, but the Amazon culture is not to just deal with it, every delay is studied, and plans made to avoid repetition, but delays do occur.

The final delivery is critical, that is the customer delight that Jeff Bezos constantly emphasizes.

To make this happen 7 days a week when the traditional delivery services worked a 5-day week, Amazon pressured its outside delivery carriers to change.  Amazon first required the USPS to add Saturday and Sunday delivery to their contract and then forced both UPS and FedEx to do the same.  That rippled throughout the industry and now all third-party logistics companies, and the carriers, have had to adapt to remain competitive.

Amazon took the cue, but moved it even further, from Uber. It now has flex drivers who bid on delivery zones as independent contractors.  They bid through an app on their phone.  The app tells them what to deliver and where.  It controls their time on task so that Amazon can give the customer truly amazing accuracy on delivery. You get those email messages now, often with a photo showing it as delivered.  Think about that, it is amazing. But, you must also understand the behind-the-scenes logistics that make it happen.

Now, Amazon has some of its own workers making deliveries, but increasingly they are the flex drivers who are not on the Amazon payroll.  Amazon owns some trucks with the logo, but increasingly finances the purchase or lease of those vehicles to the flex drivers so that they maintain their independence but still carry the Amazon name.

These drivers are often part time and work odd times while holding down other jobs.  All this works really fast.  The aim of Amazon is two-hour delivery.  It already provides 2-day.  When you think through the various levels of the logistics system it seems like an impossible task, and in areas without high population density, it is, but when you think about how Amazon groups these diverse distribution centers, it becomes possible.

An item can be pulled and packaged in 10 minutes from a Findlay area stocking warehouse.  Be delivered to Churchill in less than an hour.  Unloaded and sent to North Versailles in 30 minutes, and then delivered back to a Braddock residence in 15 minutes.   The more final delivery stations Amazon can set up in east Pittsburgh, the more efficient this becomes.

To further support this service, the AI will predict what items customers in the east of Pittsburgh will be ordering, so you will see actual storage of selected products in the proposed Churchill facility as well.  They only need the 40 ft high first level, and part of the 2nd level for the final sorts.   Amazon frequently has different roles within the same distribution center, so this is not new, just being raised to a higher level in this super-sized building.

As we have stated before, this exact blueprint is the latest generation design from Amazon and is being repeated at 12 identified constructions around the country over the last year.  In two sizes, the 640,000 and the 840,000-sf footprint, plus or minus about 20,000 sf depending on local conditions.   An Amazon spokesperson is quoted as confirming the count is actually 20 of these mega facilities.

Our exhibits give an example of specific locations for one of each size.

The outline of the building on the Hillwood PowerPoints is not that of the actual building to be constructed in Churchill.  It is a place holder design that is used on all presentations, so we have not been provided with any renderings of the actual building the applicant intends to construct.  For example, you can see this in the Alcoa exhibit R 1.

We cannot underestimate the genius of this designed facility.  The massive size is of course why Amazon and Hillwood continue to obfuscate any of these details.   Much can be gleamed from those other facilities, just like it, which have already opened up or where more details were provided to those industrial park operators.  Even now, some residents in Churchill still imagine something like the Findley facility.  This one is so much larger.

For example, exhibit R has much more detail than what Hillwood has provided to Churchill.  Such as the full dimensions of all 4 sides of the building.  That is also a   Hillwood project, with the same design as it presents to Churchill.  Plus, it is in a dedicated industrial park, not an established residential community.  Reminder, this is the same building Hillwood proposes to build in Churchill.

Hillwood knows that their project does not belong at this location, so it intentionally avoids any details relating to its actual size.  If you want to see what the actual height will be, look at exhibit V 4 before the façade is added.  It will scare you.

Now let’s also put into perspective the trucks bringing stock into this proposed mega facility.

Amazon owns the trailers with its logo, and directly or indirectly owns those without it.  Amazon has many subsidiaries and exclusive contracts.  This means it controls those trailers at every single step of the logistics system.

It does not control those warehouse-to-warehouse semi drivers who bid on a delivery.  They are like Uber drivers, or the final mile flex drivers.  They are independent contractors.  Amazon simply assigns them the trailer to deliver.

In fact, Amazon is now doing for the truck drivers what it does for the flex drivers.  It is loaning them money to go buy the rigs themselves, and then tying that loan into a requirement to only drive for Amazon.   And recently purchased rigs that it leases to those drivers with the Amazon logo.  These drivers too have deadlines to meet.  You rarely see them at the turnpike rest stops.

All the rigs coming into the Churchill facility will arrive and drop the trailer, with very few exceptions.  They will either be immediately given another trailer to carry out, or they leave the property and return when the app signals them to do so.

After dropping the load in, they can decline a new load out and drive the rig home.  Or they can go somewhere and wait until the app calls them back inside.  Think about folks at the airport waiting for passengers.  If you cannot wait at the arrivals curb, you go find a spot, anywhere you can.  In industrial zones trucks can go to the local truck stop for a coffee, or park around the area.  Since few industrial parks have 24-hour operations like Amazon, in off hours they usually park at the entrances to those closed facilities or just on the side of the road which are purpose build, wide, for semi-trucks.

Neither option exists in Churchill, nor are there any truck stops within the area.  The property could have a corral, or waiting zone, for waiting on the property.  However, the submitted drawings do not account for such a staging location on the property.  It could be added, but with every square foot already allocated for a task, what must be eliminated?    The planning commission slightly addressed this, but that is so small that it is nothing more than a token gesture.

Remember these two things.  The size of the facility, and the requirement for 2 hour and 2-day delivery.

Now we get to the real concern for Churchill.

The loading and unloading times for these trailers, and procedures, vary with the type of distribution warehouse.  Amazon knows, based on historic data and projections that some products take longer to load or unload than others.  The idea is to maximize the space within the trailer, using every single cubic foot.  Trailers have weight limits, but the products Amazon sells never reach it like coils of sheet steel would.

Documents indicate the goal within Amazon is 20 minutes pers trailer to load, and to unload.  That gives 5 minutes to park and open up and 5 minutes to close and pull away.  Workers are timed to meet these requirements. What we learned is that this supposedly applies to both loose stack and containers or pallets.  As this kind of information is usually not shared with the public, we cannot fully verify it is accurate as it is based on interviews with Amazon employees, but it is logical when one considers a goal of two trucks per hour per loading dock.  It may be wrong, and it may be correct for the Churchill proposal.  In any case, it illustrates that each loading dock turnover is no less than 24 trailers in a 24-hour period.  That is only 1 trailer each hour.

Why this is important to know will be discussed later.  But it is important to know.

We must understand the huge amount of storage space a building of this size can contain.  Remember, that the national average for a distribution center is 180,000 sf foundation, versus the 639,000-sf foundation the applicant wants to see in Churchill.

Almost 3 million sf of floor space requires a lot of trucks to fill that much space, space equivalent to the entire Monroeville Mall, less just the theatre.

Amazon standard is already 2-day delivery with Prime.  It is well on its way to 1-day delivery.  The stated goal is for many products, to have 2-hour delivery.

Think about that.  If 2-hour delivery for say 20% of the products, that means 20% comes in and then goes out immediately, if not already on hand.  Every item that goes out, needs an item to come in all within the 2-hour time frame.   Just imagine how many vehicles will be required for that rapid turnover.

We also will see some consolidation packaging on site.  Meaning some products, (all known and predicted by the AWS data) will come in loose and be combined into a single box for final delivery.     This requires more time than just rerouting a sealed box.  But it too must meet the Amazon quick turnaround requirements.

Think about the whole program.  If a single box comes from LPRZ (Maple Heights OH) shipped through CLE2 (Fulfillment center near Bedford OH), it will have already lost at least a day in transit.  Therefore, Amazon has just another 24 hours for final delivery to the customer, and still meet the current 2-day delivery obligation.

Trivia, there is a website Waze.com with a sub program that will calculate travel time specifically for Amazon drivers from one center to the next.    You can also see the driver’s complaints about drop load and pick up problems which Liz Casman illustrated in her presentation.  It also works for cars.

How will Amazon in Churchill handle a 2-hour delivery requirement it committed to in 2019?

What has this to do with the Churchill facility?

It simply means that this distribution center will be emptying and refilling almost 3 million sf of floor space of products every 24 hours.  In simple terms, the number of trucks given by Hillwood cannot support this facility.  And repeating those numbers is a complete distortion of reality.

In the October 30, 2020, memorandum to borough council, Gateway engineers estimated, based on Hillwood data, (not reality data) that there would be 4,054 daily trips generated.    That is 168 trips per hour, 24 hours a day.

That apparently includes employees, but the data below only concerns truck traffic.  You will need to add that other traffic, including vendors and employees on top of those numbers we are discussing.

How do we know how many trucks are needed to supply this facility?  Let’s look at some basics.

This is an Amazon designed facility, Hillwood is just building it.  It has 49 docks.  Every single motion at Amazon is tracked and measured for increased efficiency.  They are confident that it will handle the initial basic requirements of an almost 3 million sf of floor space.   Unless of course, they intend to add additional buildings on the property after this one is built, but personally I don’t think so, although, they have added more parking after the initial construction in many locations.

Without going into the density and average packages by weight and size, we can do some rough estimations to calculate the number of trucks that will utilize the site.

One way to look at this is to assume every truck in, leaves with a full load, which we know is not true, because they must take those emptied trailers back to be refilled again.  So, the stock out will involve different trucks, hence double the numbers.

If we use 1 trailer unloaded per hour instead of the 2 that will give us 49 docks times 24 hours or 1176 semi-trucks (the 18 wheelers) coming into the property in a given 24 hours. Then, those trucks will also need to exit the property so add another 1176 movements.

We simplify by assuming that those cabs drop a full trailer and pick up an empty one with that single entry and exit.  Which in reality does not actually happen as most cabs must leave the property and are called back later through the app to collect an empty trailer.

1176 trucks in and then 1176 trucks out will give us 2352 semi-trucks movements, not 600 or 700 in the Hillwood presentation. entering and exiting the property on local roads.  That is almost 4 times what Hillwood claims.

Even if you say there are inefficiencies of say 33%, not likely with Amazon, but to be generous, you will still have an absolute minimum of over 1500 (1552) big rigs coming and going every 24 hours. Which is more than 60 transits per hour, one every single minute of a 24-hour day.  Think about that:  One truck every minute   24 hours a day.   1 truck every minute.  This conservative number is far more than the 600 or so Hillwood projects.  That is why we call their numbers misrepresentations.

At the planning meeting the Amazon representative stated, with a map, that they will not be using semi-trucks down Beulah Road south to North Versailles.  He totally ignored the route north on Beulah Road toward Lawrenceville which has no room, whatsoever, to widen and is already encumbered at certain times of the day by semi-trucks.   Everyone in Churchill has encountered them at various stages, especially at Beulah and William Penn. What are the alternatives to supply Lawrenceville and the remaining eastern Pittsburgh locations when they come online?

As to the southern route, the reason given is that the restrictions through Turtle Creek do not allow for semi 18 wheelers.

Anyone with even slight knowledge about Amazon can see that the circuitous route given would never be tolerated by Amazon.  The solution is quite simple. Use class 6 Box Trucks.   That is what all other truckers use in urban environments for pickup and delivery (XPO, YRC, FedEx, Estes SAIA, Old Dominion…)  18 Wheelers are Class 9 or 10 due to their weight and negative impact to roads, you do not use them in urban environments unless you have to.

That Amazon spokesperson also neglected to state that those independent contractors, using GPS, and with specific delivery deadlines, are not under the control of Amazon.  They will go wherever they want to make that delivery. Let that soak in a little bit.

The elephant in the room is still Beulah Road North to Lawrenceville and the other future final mile locations we still do not know about. We already have serious delays at Frankstown Road and down to Allegheny Boulevard.  Adding several hundred more semis or several hundred cl 6 Box trucks will make it a nightmare.

The few semis trucks that currently use Beulah to go north already create disproportionate problems on those roads.  Beginning with taking the turn north on Beulah from the parkway exit.   The first turn right they come 4 feet into the left turn lane now.

Going left is even worse as it is an acute angle more that 90 degrees.  Few semis currently use this, but when they do it is dangerous.  Now image one at every sequence of light changes going south under the parkway into the property and its right turn only 120 feet on the right  as they need to merge over to that next lane.

Like so much of this applicant’s assumptions, Churchill borough simply does not have the infrastructure to support any distribution center, and especially not this mega center.

There is simply no room to widen the roads without eminent domain taking property from the old golf course, and homes up to the top of the hill north and taking property south along Beulah to Turtle Creek.

Nor do we address the increased flooding by paving over the water shed.  Others detailed those risks.  But they will affect traffic after it leaves the property in those flood prone locations.

At half the size of a semi, and half the driving time by the statement by the Amazon rep to go through Monroeville. we can calculate that for every semi bringing freight into the facility there will be two CL 6 trucks (or a full Semi to some locations) so the 1176 outbound semi-trucks becomes an additional 2352 urban box trucks.  Simply adding 1176 and 2352 we are at over 3500 potential trucks using local roads, every day, including 3 AM.  Divide that by 24 hours and see for yourself how any trucks you will hear between 3 AM and 4 AM while you try to sleep.

Now let’s talk about other vehicles.

There have been discussions about the number of parking spaces in the proposal and a reduction to that count. This is a distraction, a tweaking around the edges of what is not relevant.      Some comments indicate this as some kind of victory; it is useless.

The number of employee parking stalls did not come out of nowhere, they are what Amazon expects will be needed. Using 500 employees per shift as a benchmark, does not account for the non-amazon employees, such as security who will be coming in daily to the property.  Nor does it account for EMS or Fire or police or other contractors who will inevitably be called to the property.  Nor does it account for the twice-yearly part-time employee expansion.

Potential workers are interviewed and trained by independent employment companies on site to work at each given facility.  Most workers are not on the Amazon payroll, but that of the private employers and on contract to work at Amazon.   With close to 100% annual employee turnover, this will be a continual process on site with an unknown number of employment company employees adding to the potential interviewing workers coming and going daily.

Keep in mind, 6 weeks at Christmas and 2 weeks in July, means the total number of workers on site will double.  Yes, double.

If there are no spaces for the temp workers to park in peak seasons, where will they go?  Just like the waiting trucks, they will find a spot in the neighborhood to park.  How many spots in Churchill can you locate right now which are suitable for these truck cabs and employees to park in, or can accommodate a line of vehicles?

In addition to the trucks, as they overlap, at least 1000 of the parking spaces will be used all the time.  Not 500 as some claim, but 1000 because the new shift must arrive before the old shift punches out and they need open parking stalls.

Somewhere near 500 worker vehicles in, and 500 out at each shift change.  Add this to the number of trucks and we are facing an inevitable and unsolvable traffic problem in delays, accidents, and maintenance cost.

If this had been a typical distribution center of almost 3 million sf, it would have required 3000 parking stalls.  Because this facility is so automated, accommodating trucks is more important than accommodating individual SUVs and pick-ups.

This is Amazon with two massive surges in July and the holidays where employee count doubles for a combined 8-week period, roughly 2 months of the year.  That means each shift change needs another 1000 + spaces.  The only way this can be accomplished on site is to add parking in the already diminished green space,  or build multi-level garages.  Or, the 1000 plus vehicles will all park off site in the community.

The entrances and exits will simply not be capable of handling that dramatic increase in traffic.  There are so many bad assumptions for the interchanges on Beulah and on Greensburg Pike, it would take another 15 pages of text to point them all out.  If re striping and coordinating lights could alleviate previous rush hour traffic jams, it would have been done a long time ago.  And the amount of lane widening space at the pinch points is almost zero.  How do you widen Beulah under the parkway?

So those who have accepted the numbers from Hillwood and Amazon are ignoring the reality of this facility.

Not accounting for the actual increase in traffic leads to an even more significant delusion.  All the analyses:

The traffic studies

The sound studies

The light studies

The soot and air quality studies

The noise studies

All are based on the Hillwood traffic numbers, that are at minimum 1/3 of what will take place.  Each of these studies is based on false premises.  They should all be thrown out and the borough, not Hillwood, collect real data.

The fact that all these studies are based on selected averages, and not the extremes, such as the recent Saturday downpour, also makes these studies misleading as extreme weather becomes the new normal. For example, the floods in Germany or heat in Portland.   It is false to compare out of date historic averages to today’s reality with its increasing extremes and say that Hillwood studies for this project meet the requirements.  They do not.

Others will go into this in more detail.  And details we have.

One area overlooked is that the 60% of traffic exiting to the Greensburg Pike will be uphill in low gear, and idling uphill at the light to exit.  This was not considered in the pollution study or the noise study, which relied on trucks noise and pollution averages on flat surfaces.  Consequently, both of those requirements fail the criteria for approval.


It is tough to make good decisions.  How does one crystallize his principles to make that right decision, especially in ones like this, where there are unknown risks, and questionable data to substantiate the decisions?

Residents have spent thousands and thousands of hours generating a solid case for the borough of Churchill to reject the applicant’s proposal.  The Casman documentation recently added to the Churchill Future website has 35 pages of precise and documented reasons to vote no.  Read it, footnotes and all.   Knowledge will be greatly expanded.

Throughout this whole process Hillwood has behaved in a duplicitous manner. The statements and PowerPoints are full of false comparisons, exaggerations, misleading graphics, distortions, omissions, and distracting comments and details.  There is a reason Jeff Bezos forbids PowerPoints at Amazon, it gives the presenter too much wiggle room to hide and distort the big picture.  So just how credible is Hillwood as a developer?

Example, most people are not aware that the original proposal up to August 3rd 2020 was for a 640,000 sf facility.  Then it jumped to 840, and then back down to 639 in January, supposedly as a concession.  Look at the workshop presentation on the Churchill Borough website, it is all there.   I also added it as exhibit V 1.

An example of the 5 level 840,000 sf foundation, the same as Hillwood proposed in November, can be seen at the LogistiCenter at I-95 in Wilmington Delaware.  The Delaware Business times described it as needed for Amazon to assure its goal of one day delivery.  The proposed building outline is exactly the same as Hillwood used, although the developer is Dermody Properties, not Hillwood as at the Alcoa TN location.

Promises were made to fix omissions and faulty data later on every single one of the studies.  For clarity, there are deficiencies are on every single study submitted by Hillwood.  Hillwood cannot be trusted to fulfill any such promises based on the overwhelming false data already submitted.  Remember, those distortions are deliberate, not accidental.  This company does these developments for a living, so those distortions had to have been deliberate.  We can not trust Hillwood.  Hillwood is not credible.

Some say, what about the rights of the land owner.  It was purchased as a C 1 zoning, so it can and should be developed as C 1 property, not a C 3 or M 1 leading to destruction of our residential community.  This project is not consistent with the comprehensive plan, nor is it consistent with past development of the borough and it current character.  That alone is reason to vote no.

Voting against the Hillwood proposal will assure that the whole borough is united. A united borough is the best defense should Hillwood decide not to accept the no vote.  It should be clear to all that the majority of residents oppose this project. It should also be clear we all support the site development with businesses that complement the community, not destroy it.

Apparently, Hillwood and Amazon underestimated the level of professional and intellectual capacity of the residents and their ability to see through misleading and distorted justification for their project.  Not only does Churchill have professional academics in the fields related to this project, but it also has the financial ability to resist the project far beyond the application phase.

Hopefully, Hillwood and Amazon will accept a no vote and moves on to their plan B.  If not, be assured that the residents will not cease their involvement and resistance to Hillwood and this Amazon distribution center. The discovery phase in any legal case always uncovers something interesting.

Each council member will relate differently to each of the concerns residents have brought up in these meetings, or in something we may have referenced, but together we all will need to say no and defend the community against this destruction.

This massive distribution center is the wrong business for this location.  The risks are too high, and the damage too severe.

Given the disingenuous data supplied by the applicant, and the high probability for significantly more traffic generation than projected, the applicant has not provided what is required for the approval of this project.

We ask the council to vote no on the current Hillwood proposal as it is detrimental to the safety and to the welfare of Churchill Borough.

Murray Bilby

2424 Churchill Road