Basics — what are time windows?

A time window is the allowed arrival interval (e.g., 09:00–11:00). Combine it with service time (duration on site) and priority. Optimizers respect these rules while minimizing distance and balancing load.

Step 1 — Clean your stops

Use the fields below so dispatchers and optimizers speak the same language.
Field Meaning Example
address / lat / lng Location of the stop (geocode preferred) 25.1972, 55.2744
contact / phone Site contact for dock or guard “Ahmed – 050 123 4567”
service_time_min Expected minutes on site (unload/setup) 25
tw_start / tw_end Time window when arrival is allowed (24‑hour HH:MM) 09:00 .. 11:00
priority Urgency score (1=must hit, 3=flexible) 1
vehicle_class Allowed vehicle type for restrictions van / truck / reefer
dock_notes Special instructions (gates, low clearance) “Gate code 4412, Dock 3, 3.2m max”

Definitions: tw_start and tw_end specify the earliest and latest allowed arrival time at a stop. If a stop has 09:00–11:00 and a 25‑min service time, the optimizer will arrive between 09:00 and 11:00 and schedule departure ~25 minutes later.

Signals to monitor (first 90 days)

stop_id,address,lat,lng,tw_start,tw_end,service_time_min,priority,vehicle_class,dock_notes,contact,phone C001,Warehouse A,25.2048,55.2708,08:00,10:00,20,1,any,”Gate code #4412, Dock 1″,Ahmed,0501234567 C002,Store Downtown,25.1972,55.2744,09:00,11:00,30,2,van,”Low clearance 3.0m”,Sara,0509876543 C003,Hospital East,25.2222,55.3123,13:00,16:00,25,1,reefer,”Pharma delivery”,Dr. Noor,0501122334 2

Step 2 — Set constraints that matter

Constraint Examples Notes
Time windows 08:30–09:30; 13:00–16:00 Hard vs soft (penalty if missed)
Service time Unload 15–45 min Can depend on quantity/pallets
Driver hours Max 9h, breaks every 4.5h Regional rules apply
Vehicle capacity Weight/volume/pallets Multi‑stop load constraints
Skills/permits Hazmat, freezer, site card Match driver to stop needs
Depot & cut‑off Start/end depots; curfews Consider traffic peaks

Step 3 — 30‑60‑90 rollout

  • Day 0–30: pilot 20–30 stops/day on 3 routes; define windows & service times; validate travel matrix; baseline KPIs.
  • Day 31–60: expand to all routes; add soft window penalties; integrate fuel prices & traffic; start exception queue.
  • Day 61–90: multi‑depot optimization; add detention rules; weekly review of missed windows & overloads.

KPIs that prove savings

  • On‑time delivery % (target: > 97%)
  • Distance per stop (target: ↓ 10–20%)
  • Routes per day / vehicles used (balance)
  • Detention hours (target: ↓ 30%)
  • First‑attempt success % (target: ↑)

💬 FAQs

Start with soft windows (e.g., 09:00–12:00) and tighten based on history. Use priority for critical stops.
Daily for scheduled routes; intraday when exceptions happen (no‑show, breakdown, urgent add‑on).
No—good travel time estimates plus reliable service times deliver most savings; add live traffic for intraday replans.
Use precedence constraints (pickup before delivery) and capacity rules by pallet/volume so loads never exceed vehicle limits.
Yes—set service areas or driver‑stop affinities to respect customer relationships while still optimizing within each territory.
Start with a default per stop, then add multipliers by quantity, dock type, or customer class. Tune weekly from actuals.
Tip: put missed windows and detention into an exception queue with root‑cause notes. You’ll find fixes fast.