Sort driver messages: failed deliveries, delays, breakdowns
Reads each driver message for what happened, whether to tell the customer or dispatch, and the charge it supports. It describes the load, never the driver.
Try it on this example
Load or stop status from the TMS: Multi-drop run 7, stop 14 of 22: Unit 6, Brookfield Trade Park. Arrived 11:52.
Driver's message
- What is the driver reporting in this message?Delivery or collection failed97%
- Why does the driver say the delivery or collection could not be completed?Access problem99%
- Does the driver's message suggest a load or delivery will arrive later than planned?Yes99%
- Does the driver's message mention an injury, a crash, an unsafe load, a spill, a fire or danger to the driver?No90%
- Does the driver need something from dispatch to carry on or finish the job?Yes97%
- Should the shipper, receiver or recipient be told about what the driver reports?Yes95%
- Could the customer fix the problem the driver reports by giving information, such as an entry code, a corrected address or a safe place?Yes94%
- Does the driver mention arrival, appointment or waiting times at a site?No61%
- Which extra charge to the customer does the driver's message support, if any?Redelivery77%
- Is the driver's message clear enough to tell what happened?Yes97%
These are real answers stored from one run on this example.
The prism behind it
Sort driver messages: failed deliveries, delays, breakdowns
Fields
- Load or stop status from the TMS
- Driver's message
Context
We are Aldous Haulage, a regional carrier running full loads, part loads and multi-drop van deliveries for shippers and retailers. Drivers message dispatch from the cab app or by text: arrivals, delays, waits at docks, failed deliveries, breakdowns and problems with the load. Each message is read here as it arrives, with the load or stop status from the transport management system (TMS). Code sorts the dispatcher's screen, sends the customer a status update, books a redelivery, and starts any waiting-time charge from the TMS timestamps. Safety messages go to a person on the safety line at once. Driver shorthand: appt (appointment), DC (distribution centre), door (loading dock door), plt (pallet), drop or stop (a delivery address on the run), POD (proof of delivery), "my hours", "my 14" or "my tacho" (the driver's legal driving time is running out), lumper (paid unloading help). The answers describe the load and the delivery, never the driver, and are not used to score or rank drivers. Driving hours are worked out by code from the vehicle logging device, not from the message.
Questions
What is the driver reporting in this message? Choice
Read the message with the load status. When the message reports more than one event, pick the first that applies in this order: accident or incident, breakdown, load problem, delivery or collection failed, hours running out, waiting at a site, running late, paperwork problem, on schedule, not about the load.
Why does the driver say the delivery or collection could not be completed? Choice
Judge from what the driver says happened, not from any code in the load status. When more than one reason is given, pick the one that stopped the delivery first.
Does the driver's message suggest a load or delivery will arrive later than planned? Yes / No
Count delays, waits, breakdowns, failed deliveries that need a new attempt, and knock-on delays to later stops. Yes: At least one load or delivery will be later than planned. No: Nothing in the message suggests a delay.
Does the driver's message mention an injury, a crash, an unsafe load, a spill, a fire or danger to the driver? Yes / No
This sends the message to a person on the safety line at once. It never replaces a call to the emergency services. Yes: The message mentions any of these. No: The message mentions none of these.
Does the driver need something from dispatch to carry on or finish the job? Yes / No
Count a request for instructions, a code, a contact, a new appointment, authorisation, recovery or a replacement vehicle. Yes: The driver asks for or clearly needs something from dispatch. No: The message is information only.
Should the shipper, receiver or recipient be told about what the driver reports? Yes / No
Count anything that changes when or whether the goods arrive, or that needs the customer to do something. Yes: The customer should be told. No: Nothing in the message affects the customer.
Could the customer fix the problem the driver reports by giving information, such as an entry code, a corrected address or a safe place? Yes / No
Yes: The problem would be solved by information or a choice the customer can give. No: The problem is not one the customer can fix with information, or the message reports no problem.
Does the driver mention arrival, appointment or waiting times at a site? Yes / No
Count clock times, appointment times and how long the driver has waited at a site. Code reads the times from the TMS; this only flags the message for billing. Read the driver's message only; the load status times come from the TMS. A delay on the road, or a slot at a later stop, is not a time at a site. Yes: The message mentions at least one of these times at a site. No: The message mentions no such time.
Which extra charge to the customer does the driver's message support, if any? Choice
Judge from what happened, not from whether the customer will be billed; billing decides that with the TMS times.
Is the driver's message clear enough to tell what happened? Yes / No
Shorthand, abbreviations and poor spelling are normal and still readable. Yes: The message can be read and says what happened. No: The message is garbled, cut off, or too short to tell what happened, such as "ok" or "call me".
Lens columns
event, event_probability, failed_delivery_reason, failed_delivery_reason_probability, eta_affected, eta_affected_probability, safety_concern, safety_concern_probability, needs_action, needs_action_probability, tell_customer, tell_customer_probability, customer_can_fix, customer_can_fix_probability, time_at_facility_mentioned, time_at_facility_mentioned_probability, accessorial, accessorial_probability, text_judgeable, text_judgeable_probability
Run it on your own text
Add this prism in the app, change any question, and test it on a file of your own.