Routing & Route Planning Engines
A transport manager overseeing 40 vehicles across the Paris region receives between 180 and 250 deliveries to organise each morning. Two-hour time slots demanded by clients, low emission zones which are changing the rules of the game, temporary drivers unfamiliar with the areas, refrigerated lorries which cannot transport everything. By 6:30 AM, the routes must be ready.
At 10:00 AM, one client cancels whilst another rings to bring forward their delivery. At 2:00 PM, a vehicle breaks down. The morning schedule no longer exists. Operations teams spend their days firefighting rather than optimising.
Drivers ring because the address is incorrect, the entry code is missing, the client is absent. Meanwhile, last-mile costs represent 40 to 50% of total transport costs. And margins are eroded when fuel prices increase by 15% in six months.
This solution belongs to the Business Operations & ERP family.
Challenges holding back your performance
Excel with macros developed eight years ago by someone who is no longer there. A TMS that performs static planning but cannot recalculate in real time. Google Maps to verify routes one by one. Drivers who optimise their own rounds because they know the terrain better than the software.Generic tools ignore your business constraints: this vehicle cannot deliver to that customer, this driver has electrical certification, this delivery slot depends on an appointment made the day before.Result: empty kilometres, avoidable depot returns, delay penalties, and an operations team manually compensating for the system’s shortcomings.
Our Technical Approach
We begin with your actual data: delivery history, service times observed by customer type, average speeds by zone and time slot. Not theoretical averages. The VRP engine incorporates your real constraints: vehicle-goods compatibility, driver skills, contractual delivery windows, statutory break times.
The geocoding is enhanced with field information that your teams have accumulated over years: car park entrances, loading bays, time slots when particular hypermarkets accept lorries. The driver application does not simply display a route: it feeds back information in real time so that operations can reassign a delivery in under two minutes when an unforeseen event occurs.
Electronic proof of delivery closes the loop: photograph, signature, comment, GPS time stamp. This enables analysis of variances and improvement of future planning.
Geocoding & Matrices
Transformation of textual addresses into GPS coordinates and calculation of distance matrices.
VRP Engines
Resolution of the Vehicle Routing Problem with advanced heuristics in very short time.
Driver App
Integrated navigation, display of unloading order and parcel scanning.
Proof of Delivery (ePOD)
Electronic signature capture, photographic proof of delivery and certified time stamping.
Dynamic Recalculation
Live route adjustment in the event of customer absence or traffic incident.
Vehicle Constraints
Compliance with clearances (bridges, LEZ zones) and refrigeration capacities.
Technical Architecture
Enter once, let the rules schedule, read a single view: this foundation is shared across the family — you will find it in Order Management System (OMS) and in Workshop & After-Sales Management.
- Geocoding & MatricesTransformation of textual addresses into GPS coordinates and calculation of distance matrices.
- VRP EnginesResolution of the Vehicle Routing Problem with advanced heuristics in very short time.
- Driver AppIntegrated navigation, display of unloading order and parcel scanning.
- Proof of Delivery (ePOD)Electronic signature capture, photographic proof of delivery and certified time stamping.
- Dynamic RecalculationLive route adjustment in the event of customer absence or traffic incident.
- Vehicle ConstraintsCompliance with clearances (bridges, LEZ zones) and refrigeration capacities.
Smooth and frictionless integration
Your TMS remains the master system. The routing engine connects to it to retrieve transport orders and return calculated rounds. Your customer, vehicle and driver repositories are not duplicated. We interface with your existing telematics for real-time vehicle positioning.
Progressive deployment possible: a pilot depot, a flow typology, before generalisation. Drivers keep their usual telephone, the application installs in five minutes.
Audit of flows and constraints
Analysis of your current rounds, identification of undocumented business constraints, measurement of discrepancies between planned and actual delivery.
Configuration of the engine
Configuration of calculation rules according to your actual constraints: vehicles, drivers, customers, regulations. Testing on historical data.
Field Operator
Deployment on a restricted perimeter with daily support. Adjustments based on feedback from drivers and operations.
Generalization and autonomy
Progressive extension to other workflows. Training of teams in indicator analysis for continuous improvement without external dependency.
Measurable results for your organisation
- Reduction of 12 to 18% in kilometres travelled on multi-point routes
- Daily planning time reduced by a third for the operations team
- Customer slot compliance rates increasing from 78% to 94%
- 60% reduction in driver calls for missing information
- Complete traceability of deliveries with actionable evidence in case of dispute
- Ability to reassign a delivery within 2 minutes in the event of unforeseen circumstances
Clarifying your decision-making
Our drivers are not comfortable with digital tools, how does this work?
The driver application is designed to be used with gloves, in full sunlight, by someone who has forty-five seconds between two deliveries. Three buttons per screen at most, and training in the field rather than in a classroom. Drivers who are reluctant at first often become the system’s strongest advocates, once they see they are no longer called to report their position.
We already have a TMS, are you going to ask us to change it?
The routing engine complements your TMS rather than replacing it. Clarendis adapts to what you have, whether a market product or an in-house development. The aim is to add the missing computational intelligence, not to challenge your processes.