Reconciliation read

Planned, measured, billed. The gap, with its denominator.

The seller's system holds what was planned. The position trace holds what was driven. This is the ledger between the two, set beside what was actually charged, and it states how much of the window it could read before it states anything else.

Talk about a deal Open the ledger

Reconciliation read

OTR trucking Flat service Additive service Rail carload

Mon 15 Jun – Sun 21 Jun 2026

Modeled sample data

Book

OTR truckingvehicle-hour
Flat servicecrew-hour
Additive servicecrew-hour
Rail carloadcar-day

Window

Range: Mon 15 Jun – Sun 21 Jun 2026
Basis: 120 truck-days, 14 h duty clock; 30 crew-days, 07:00 to 16:00 shift; 25 crew-days, 8.5 h paid shift; 180 car holds, placement to release
Scope: 325 stops, 48,141 mi; 301 services, 420 accounts; 160 visits, 47 extras, 260 accounts; 70 chargeable car-days

One window. This read is a rendered artifact, so the range is stated rather than offered as a control that would have nothing else to select.

Time in state

Moving55.5%
On dock38.3%
of which idling13.5%
of which detained8.9%
Not placed6.2%

Shares of the duty clock. Idling is engine-on and stationary, so it sits inside the dock time rather than beside it. Against engine hours alone it reads 19.6%.

On site49.8%
of which off book0.9%
Driving28.4%
Break8.0%
Unaccounted14.2%

Shares of the paid shift. There is no engine feed on this book, so unaccounted time is both the idle proxy and the coverage gap. The read says so rather than picking one.

On site66.3%
of which unbilled6.6%
Driving18.6%
Unaccounted15.2%

Shares of the paid shift. Unbilled extras are a state and not a rounding: crews were on a property doing work that never reached an invoice.

A placed car has no engine and does not move under its own power, so there is no moving, idling or stopped split to report on this book. The whole read is one clock against one published tariff.

Exceptions

All Priced Not priced

Not priced is not the same as clean. It is a finding the engine measured and refused to attach a number to, and it stays in the ledger.

Detention on this book billed $9,513.87 against $9,952.85 of operating cost on the same 149 hours, a $438.98 shortfall, read across the 93.8% of duty hours the position trace could place.

ReconciliationModeled

Stops served

Planned: 329 dispatched
Measured: 325 made
Billed: no charge
Variance: 4 stops not made

P5 route completion · manifest reconcile against the dispatch plan

Detention, hours past free time

Planned: 0 h chargeable
Measured: 149 h on 121 of 325
Billed: $9,513.87
Variance: −$438.98

P3 detention · free_time_s=7200, rate=per_hour, $63.71/h recovered against $66.65/h of cost

Miles off corridor

Planned: 0 mi
Measured: 718 of 784 mi
Billed: $0 recovered
Variance: $1,680.09 carried

P6 off-route · sustained deviations only, persistence_s=300, priced at $2.34 per mile

Idle, engine on and stationary

Planned: 0 h
Measured: 19.6% of engine hours
Billed: $0 recovered
Variance: $724.27 burned

P7 utilization · idle_threshold_s=300, 186 gal at 0.82 gal/h, priced by the engine's two-rate idle burn

Bar is measured. Tick is billed. Caret is planned. The rule under the bar is the span between what happened and what was charged, and hatching is quantity the read could not determine.

CoverageModeled

Stated before the rates above, not after them. A rate without a denominator is an assertion.

93.8% of the 1,680 duty hours in this window are placed by the position trace. The remaining 6.2%, roughly 104 hours, carries no segment at all. It is excluded from every rate above rather than assumed idle.

Fired
5 of 9 KPIs. P3 detention, P5 route completion, P6 off-route, P7 utilization, P8 speeding.
Did not fire
4, each with a reason. P1 dwell, because over-the-road uses P3 for the same physical fact with the other polarity. P2 unauthorized stop, because this trace has no unknown-class stops to fire on. P4 after-hours, because no authorized-window schedule is modeled here. P9 perimeter, because the trace holds no geofence exit events. A blank is an unfired KPI, never a clean score.
Not determinable
25 events. Brief off-corridor excursions that did not clear the 300-second persistence floor, carrying 66 of the 784 observed deviation miles. They are reported as uncounted, with the count, and never as zero.
Not priced
P8 speeding, 8 of 329 moves. The threshold is absolute while the published band is measured against posted limits, so the two are not the same measurement. The row stays in the ledger and carries no dollar.
Basis
Idle here is measured against engine hours from a real engine feed, so it is comparable to a telematics idle figure. On the two service books it is a proxy against the paid shift and is discounted on their own coverage panels.

Exceptions, by dollar impactModeled

$1,680.09 Miles carried off corridor P6 off-route · 718 mi sustained on 53 of 329 moves

Modeled evidence. One row from this run, not a client record.

The one panel on this page where a map earns its place. The question is genuinely spatial: the corridor is the plan, and the miles were counted against the departure from it.

evidence
EvidenceRef::OffRoute

segment_index
18

start_ts
2026-06-17 11:04:22Z

end_ts
2026-06-17 11:39:05Z

off_route_m
21,140.00

charge
$30.74 at $2.34 per mile

Rule P6 off-route, sustained deviations only, persistence_s=300. A deviation that does not hold for 300 seconds is suppressed and counted as not determinable, never as zero. The mileage ledger behind this row is fleet-cleaning grade, not survey grade, and the read says so before the number is used.
$724.27 Fuel burned at idle P7 utilization · 186 gal, 19.6% of engine hours, on 120 of 120 truck-days

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Idle

idle_seconds_total
7,020

engine_on_seconds
35,820

burn
0.82 gal/h, inside Argonne's measured 0.64 to 1.00 range

charge
$6.23 on this truck-day

Rule P7 utilization, idle_threshold_s=300. Idle is engine on and stationary within the stop radius, so it is not the complement of moving time and the two are not summed. The denominator on this book is engine hours, taken from a real engine feed.
$438.98 Detention recovered below its own cost P3 detention · 149 h on 121 of 325 stops, billed $9,513.87 against $9,952.85 of cost

Modeled evidence. One row from this run, not a client record.

No map in this panel. The finding is how long the truck sat, not where it sat, and the free-time clock is the whole of it.

evidence
EvidenceRef::Detention

segment_index
41

geofence_id
L-0287

geofence_class
customer

arrival_ts
2026-06-18 08:12:00Z

dwell_seconds
15,720

detention_seconds
8,520

charge
$150.78 at $63.71 per hour

Rule P3 detention, free_time_s=7200, rate=per_hour. This stop is also one of the 19 that ran past four hours. The comparison figure is computed the same way the engine computed the billed side, per detained stop with half-away rounding, so the shortfall is not a rounding artifact.
Not priced Stops dispatched and not made P5 route completion · 4 of 329, completion 98.8%

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::ManifestMissed

stop_id
LD-1174

geofence_id
L-0433

charge
none

Rule P5 route completion, manifest reconcile against the dispatch plan. The stop is in the manifest and absent from the segment stream, so this is a reconciled absence rather than an estimate from a miss rate. A miss cites only the manifest identity, because there is no observed segment to point at. This book's tariff carries no charge for a stop that was not made, so the row carries no dollar.
Not priced Moves over the absolute speed threshold P8 speeding · 8 of 329 moves, 2.4%

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Speeding

segment_index
7

start_ts
2026-06-16 14:22:09Z

end_ts
2026-06-16 15:01:47Z

max_speed
32.104

charge
none

Rule P8 speeding, threshold_mps=31.29. The threshold is absolute and the published band is measured against posted limits, so the two are not one measurement and must not be presented as one. The read is directional and the row carries no dollar. There is no driver score anywhere in this read.

This book holds 316 visits on contract and the crews made 301, leaving $1,191.39 charged against work that was never delivered, read across the 85.8% of the paid shift this trace accounts for.

ReconciliationModeled

Scheduled visits

Planned: 316 designed
Measured: 301 serviced
Billed: 316 on contract
Variance: 15 visits, $1,191.39

P5 route completion · manifest reconcile against the designed route, priced at the book's own per-visit rate

Accounts on the book

Planned: 420 on the book
Measured: 403 ever scheduled
Billed: 420
Variance: 17 never scheduled

Designed against driven · an account with no scheduled visit anywhere in the window, cross-checked against the visit history

Stops served off the book

Planned: 0
Measured: 9 served
Billed: 4 flagged, $134.45
Variance: $75.85 under the floor

P2 unauthorized stop · min_dwell_s=900. Five of the nine fell under the floor, so total off-book labour is $210.30 against $134.45 captured

Jobs running long

Planned: 0 overage
Measured: 299 min on 15 of 301
Billed: flat price, absorbed
Variance: $419.14 of crew time

P1 dwell · threshold_s=3600, roles.whitelist=[known_customer], priced by the engine at $84.00 per crew-hour

Unaccounted shift time

Planned: 0
Measured: 14.2% on 25 of 30 crew-days
Billed: paid regardless
Variance: $3,221.10

P7 utilization · idle_threshold_s=300. A proxy, not an engine fact: the denominator is the paid shift, so this is not comparable to a telematics idle percentage

Bar is measured. Tick is billed. Caret is planned. The rule under the bar is the span between what happened and what was charged, and hatching is quantity the read could not determine.

CoverageModeled

Stated before the rates above, not after them. A rate without a denominator is an assertion.

85.8% of the paid shift on this book is placed into a measured state. The remaining 14.2% is unaccounted, and on a book with no engine feed that span is both the idle proxy and the coverage gap at once. It is reported as unaccounted rather than resolved into one of the two.

Fired
4 of 9 KPIs. P1 dwell, P2 unauthorized stop, P5 route completion, P7 utilization.
Did not fire
5, each with a reason. P3 detention, because no customer-side free-time clock exists in grounds maintenance. P4 after-hours, because all modeled movement sits inside the shift, which is not evidence of a clean fleet. P6 off-route, because no planned corridor is modeled and there is nothing to deviate from. P8 speeding, because the trace holds no per-hop speed. P9 perimeter, because it holds no geofence exit events.
Not determinable
5 stops. Off-book stops that fell under the 15-minute dwell floor and did not fire. That is a threshold choice hiding real unbilled work, $75.85 of it, and it is stated as a choice rather than reported as a clean result.
Basis
Idle on this book is a proxy. There is no engine-on signal in the trace, so the denominator handed to P7 is the paid shift and not engine hours. It is a weaker fact than the trucking book's and is not comparable to a telematics idle percentage.

Exceptions, by dollar impactModeled

$3,221.10 Paid shift time with no work attached P7 utilization · 14.2% on 25 of 30 crew-days, proxy basis

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Idle

idle_seconds_total
4,601

engine_on_seconds
32,400

basis
paid shift, not engine hours

charge
$107.36 on this crew-day

Rule P7 utilization, idle_threshold_s=300. This book has no engine-on signal, so what P7 receives is unaccounted paid shift time and the denominator is the shift. The row is reported at proxy grade and the coverage panel above discounts it accordingly.
$1,191.39 Charged for visits that were never made P5 route completion · 15 designed stops absent from the trace, completion 95.3%

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::ManifestMissed

stop_id
SV-0912

geofence_id
C-1188

charge
$79.43, the book's own per-visit price

Rule P5 route completion, manifest reconcile against the designed route. The stop is in the manifest and absent from the segment stream. On a flat-price book the contract charges for it either way, which is why the charge column here is contracted revenue for work not delivered and not a penalty.
$419.14 Crew time absorbed on jobs running long P1 dwell · 299 overage minutes on 15 of 301 services

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Dwell

segment_index
12

geofence_id
C-0341

geofence_class
known_customer

arrival_ts
2026-06-16 09:47:00Z

dwell_seconds
4,860

charge
$29.40 at $84.00 per crew-hour

Rule P1 dwell, threshold_s=3600, roles.whitelist=[known_customer]. The rate is $28.00 an hour across a crew of three, and the charge is the sum of the processor's own amounts rather than a figure this page multiplied out.
$210.30 Work delivered off the book and never charged P2 unauthorized stop · 9 stops served, 4 flagged, $134.45 captured

Modeled evidence. One row from this run, not a client record.

A map here because the finding is spatial: the crew stopped and worked somewhere that is on no customer geofence in the book at all.

evidence
EvidenceRef::Stop

segment_index
26

geofence_id
null

geofence_class
unknown

arrival_ts
2026-06-19 13:26:00Z

dwell_seconds
2,460

charge
$57.40 of labour, not invoiced

Rule P2 unauthorized stop, min_dwell_s=900, alertable=[unknown]. Same primitive as P1 with the opposite polarity and a separate class-role map, priced at the book's own $84.00 per crew-hour. Five further off-book stops sat under the floor and did not fire, so $75.85 of real unbilled work is outside the captured figure. That is a threshold choice, and it is on the coverage panel rather than in a footnote.
Not priced Accounts on the book that were never scheduled Designed against driven · 17 of 420 accounts, no visit anywhere in the window

Modeled evidence. One row from this run, not a client record.

evidence
designed plan, cross-checked against the visit history

accounts
17 of 420

scheduled visits
0

charge
none in this read

Rule An account carried on the book with no scheduled visit in the window. The read stops there. What is invoiced against those accounts is in the billing system, not in the position data, so this row names the accounts and refuses to price them. A page that priced them would be modeling the invoice rather than measuring it. There were also 107 stops served out of the planned sequence, which is a routing observation and not a charge.

Crews performed 47 extras on this book and 32 reached an invoice, leaving $3,735.00 of work delivered and never billed on $783.89 of labour already spent, read across the 84.8% of the paid shift this trace accounts for.

ReconciliationModeled

Base visits

Planned: 164 dispatched
Measured: 160 made
Billed: 160
Variance: 4 not reached

P5 route completion · manifest reconcile against the dispatch board

Extras performed

Planned: found on site
Measured: 47 performed
Billed: 32 invoiced
Variance: 15 never billed, $3,735.00

P2 unauthorized stop · min_dwell_s=900, alertable=[upsell_unbilled]. Capture came out at 68.1% on 47 extras against a modeled parameter of 63.0%

Labour on unbilled extras

Planned: 0
Measured: 6.6% of crew-hours
Billed: nothing
Variance: $783.89 spent

Effort split against the same engine counters · time on a property doing work that never reached an invoice

Extras running long

Planned: 0 overage
Measured: 3 of 47 past 90 min
Billed: absorbed
Variance: $47.93

P1 dwell · threshold_s=5400, whitelist=[upsell_billed, upsell_unbilled]. The base visit cannot trip this row by construction

Unaccounted shift time

Planned: 0
Measured: 15.2% on 24 of 25 crew-days
Billed: paid regardless
Variance: not priced on this book

P7 utilization · idle_threshold_s=300. Measured and not an artifact: the board schedules a drawn number of jobs against a fixed 8.5-hour shift, so some days finish early and some overrun

Bar is measured. Tick is billed. Caret is planned. The rule under the bar is the span between what happened and what was charged, and hatching is quantity the read could not determine.

CoverageModeled

Stated before the rates above, not after them. A rate without a denominator is an assertion.

84.8% of the paid shift on this book is placed into a measured state. The remaining 15.2% is unaccounted, and with no engine feed here that span is both the idle proxy and the coverage gap. Idle minutes per crew-day ranged from 4 to 266, which is why an average would have been the wrong figure to publish.

Fired
4 of 9 KPIs. P1 dwell, P2 unauthorized stop, P5 route completion, P7 utilization.
Did not fire
5, each with a reason. P3 detention, because there is no customer-side free-time clock in this vertical. P4 after-hours, because movement is confined to the shift. P6 off-route, because no planned corridor is modeled. P8 speeding, because the trace holds no per-hop speed. P9 perimeter, because it holds no geofence exit events.
Not determinable
None on this sample, which is a measurement and not a clean bill. Every unbilled extra cleared the 15-minute dwell floor, so the floor hid nothing here. On a book with shorter extras it would, and the count would appear on this line.
Sample
A week is a thin sample for a capture rate. The observed 68.1% is one realization of a modeled 63.0% parameter, so the parameter and the finding are what to quote, never the last digit of the observed rate.
Basis
Idle on this book is a proxy against the paid shift, because a service van carries no engine feed here. It is not comparable to a telematics idle percentage.

Exceptions, by dollar impactModeled

$3,735.00 Work delivered and never billed P2 unauthorized stop · 15 of 47 performed extras never reached an invoice

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Stop

segment_index
9

geofence_id
A-0448

geofence_class
upsell_unbilled

arrival_ts
2026-06-17 10:05:00Z

dwell_seconds
3,180

charge
$49.47 of labour, not invoiced

Rule P2 unauthorized stop, min_dwell_s=900, alertable=[upsell_unbilled]. Labour is priced at $56.00 per crew-hour, $28.00 across a crew of two. The leak concentrates by service: storm cleanup $2,050.00, shrub pruning $720.00, mulch refresh $555.00, weed treatment $285.00, irrigation check $125.00. That is 37.4% of upsell revenue on this book.
$783.89 Labour already spent on that unbilled work Effort split · 6.6% of available crew-hours

Modeled evidence. One row from this run, not a client record.

evidence
engine counters, effort split

share
6.6% of available crew-hours

extras
15 of 47

charge
$783.89 at the BLS wage basis

Rule The cost side of the same 15 extras. It is separated from the revenue row deliberately: the revenue is what a buyer could recover, and the labour is what the seller already paid whether or not anyone recovers it. Summing the two would double count the same event.
$47.93 Extras running past ninety minutes P1 dwell · 3 of 47 performed extras, 6.4%

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::Dwell

segment_index
15

geofence_id
A-0203

geofence_class
upsell_billed

arrival_ts
2026-06-18 14:11:00Z

dwell_seconds
6,600

charge
$18.67 of overage

Rule P1 dwell, threshold_s=5400, whitelist=[upsell_billed, upsell_unbilled]. The class-role map whitelists only the two upsell classes, so the denominator is every performed extra and nothing else, and the overage is priced at the same $56.00 per crew-hour. The base visit never trips this row and cannot.
Not priced Visits the crew could not reach P5 route completion · 4 of 164, completion 97.6%

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::ManifestMissed

stop_id
AV-0071

geofence_id
A-0119

charge
none

Rule P5 route completion, manifest reconcile against the dispatch board. A job the crew could not reach before the shift ended is in the manifest and absent from the stream, so the miss is reconciled rather than estimated. What that costs depends on the contract, which is in the billing system and not in this data, so the row carries no dollar.

Priced against the published per-day tariff these 180 car holds owe $8,800.00. Priced at a linear per-hour rate they compute $11,909.33, an overstatement of $3,109.33, and the linear rate agrees with the tariff on none of the 91 cases that carry a charge. That is the one question of nine this record can answer.

ReconciliationModeled

Chargeable car-days, at tariff

Planned: 0 past free time
Measured: 70 car-days on 52 of 180
Billed: $8,800.00
Variance: correct by construction

P3 detention · free_time_s=172800, rate=per_day_step, exclude_weekends=true. Free time counted in whole days, a partial chargeable day billed as a full one

The same holds, priced linearly

Planned: 0 past free time
Measured: 91 of 180 carry a charge
Billed: $11,909.33
Variance: $3,109.33 over, 35.3%

P3 detention · same population, same free time, rate=per_hour at each tier's nearest integer hourly expression of its daily rate

One scale, the linear total. The bar is the tariff-correct charge and the tick is what the hourly rate computes, so the span between them is the mispricing rather than a difference of opinion about the facts.

CoverageModeled

Stated before the rates above, not after them. A rate without a denominator is an assertion.

1 of 9 KPIs fired on this book. A placement and release record is a narrow instrument: it answers whether the charge is right and it answers nothing else, and the bar above is the honest picture of that. Eight KPIs did not fire and each carries a reason.

Fired
P3 detention, in both rate modes. Run twice over one population under two configs, which is what makes the divergence below a measurement rather than an argument.
Did not fire
8, each with a reason. P1 dwell, because a car hold is a dwell question and P3 is the dwell KPI with a price attached. P2 unauthorized stop, because every hold is at a known place. P4 after-hours, because a placed car does not move. P5 route completion and P6 off-route, because a car on a siding has no manifest and the shipper sees no planned corridor. P7 utilization, because a freight car has no engine. P8 speeding, because line-haul speed is the carrier's fact and not the shipper's. P9 perimeter, because a placement record holds no geofence exit events.
Not determinable
None. This book is not regenerated: it replays a 180-case reference corpus that was authored by an independent day-walk before the per-day step mode existed, so the figures are exact rather than sampled. 89 of the 180 holds carry no charge at all, which is a result and not a gap.
Bands
No published band exists for a demurrage rate, so nothing here is scored against one. The mechanism is documented in Norfolk Southern tariff 6004-D, Railinc RIC 6004 and Surface Transportation Board reporting. The per-day rate levels are trade press and are marked as the weaker source they are.

Exceptions, by dollar impactModeled

$3,109.33 A day-step tariff priced as an hourly rate P3 detention · 91 cases carry a charge, the linear rate agrees on 0, overstates 59, understates 32

Modeled evidence. One row from this run, not a client record.

No map. The finding is a calendar, and the calendar is the reason no hourly price can express this charge.

evidence
EvidenceRef::DetentionDays

segment_index
0

geofence_id
CAR-04417

geofence_class
customer

arrival_ts
2026-06-19 09:15:00Z

dwell_seconds
325,500

chargeable_days
1

Rule P3 detention, rate=per_day_step, exclude_weekends=true. The error has three independent sources and only the first runs one way. Whole-day billing means a three-hour overage bills a full day. Free time is counted in whole days, so the boundary sits on the day grid and not on the second. Weekends do not accrue against free time, which makes an identical hold cost a different amount depending on the weekday it began. The third source makes the charge not a function of duration alone, and an hourly rate is a function of duration alone, so no price fixes it. That the linear rate both overstates and understates on one population is the visible signature: a scale error can only go one way.
Not priced Holds that carry no charge P3 detention · 89 of 180 holds cleared free time under the tariff

Modeled evidence. One row from this run, not a client record.

evidence
EvidenceRef::DetentionDays

population
180 holds

chargeable
52 under the step, 91 under the linear rate

charge
none on these holds

Rule P3 detention, free_time_s=172800, rate=per_day_step. Half of the difference between the two columns is not a price at all: the linear rate fires on 91 holds where the step fires on 52, so it invents 39 charges the tariff does not carry. A row that reads zero here is a hold that cleared its free time, which is a measurement and not an absence of data.

Two visuals, one selection

Where in the terminal area the money went

The same week, the same book, drawn twice. Select a zone on the map and the ledger beside it filters to that zone. Select a zone or a rule in the ledger and the map redraws to match. Both are views of one selection, and either one clears it.

Corridor read

OTR trucking · Newark–Elizabeth terminal area

Modeled sample data

Data as of Sun 21 Jun 2026 23:59 UTC

Stop zonesModeled

Circle area is the dollars that zone carries. It is drawn on our own basemap, from our own tile archive.

This map needs JavaScript. The zone figures beside it do not, and they are the same numbers.

Dollars carried, by zoneModeled

Bar length is the zone total. Within it, off corridor, idle and detention in that order.

Kearny–Meadowlands

$170.71

14 stops · 1 dispatched and not made
Port Newark–Elizabeth

$158.89

17 stops
Linden–Carteret

$122.06

11 stops
Jersey City–Bayonne

$117.68

9 stops
Secaucus–North Bergen

$72.98

6 stops · 1 dispatched and not made
Newark Airport south

$57.00

4 stops

Selected $699.32 all six zones, of the book's $2,843.34

Off corridor is P6, idle is P7, detention is P3, in that order along every bar. P5 route completion carries no dollar on this book, so the two stops dispatched and not made are counted on their zones and priced nowhere.

61 of the 325 stops in this window fall inside this terminal area and carry $699.32 of the book's $2,843.34. Zone figures sum to that subtotal exactly. They are a part of the read above, never a restatement of it, and like every figure on this page they are modeled.

How to read this

What the ledger is doing, and what it deliberately is not

Denominator first
The coverage panel sits above the exception list on every book, because a rate whose denominator is unstated is an assertion. On the trucking book the coverage figure and the utilization figure are the same number, and that is not a coincidence: an hour the trace cannot place is an hour we do not count as worked.
Abstention is a result
Rows marked not determinable or not priced stay in the ledger with their counts. A KPI that did not fire is an unfired KPI and never a clean score, and a threshold that hid work says how much work it hid.
No live map
There is no live vehicle position on this screen, by design. Live position answers where a truck is now. This page answers what already happened and whether it was charged correctly, so a map appears only where the question is genuinely spatial: inside an evidence panel, and in the corridor read above, which draws a closed week rather than a moving fleet. Two of the evidence panels say plainly that they do not need one.
Precision is capped
Money is exact because the engine holds it in integer cents and every dollar is the sum of the processor's own charges. Physical quantities are not carried past what a ping cadence supports, which is why there is no score, no index, and no figure invented to fill a column.
Same engine, four books
The four books share no operating model. An industry here is a configuration file, not a code path, so the same processors run over all four and a threshold change moves every book at once. Ratios cross the books. Absolute figures do not, and they are never summed across them.

Standing

Where the numbers come from

Published figures are attributed and never rehosted. Everything else is modeled output from our own generators, run against those published bands, and it is labelled inside the ledger chrome rather than under it.

Published bands and figures are attributed to the American Transportation Research Institute, the Bureau of Labor Statistics, the Surface Transportation Board, Class I published tariffs, Argonne National Laboratory and the National Association of Landscape Professionals. Those figures are cited, not reproduced. The rail mechanism is Norfolk Southern tariff 6004-D, Railinc RIC 6004 and Surface Transportation Board reporting. Everything else on this page is modeled output from our own engines. It is not a client result, no client engagement stands behind it, and no client data was used.

Send us a book you are looking at. We will tell you what the position data can and cannot settle before anyone commits to anything.

Talk about a deal