How do I set up route optimization for recurring weekly stops?

Start with the handoff, not the map

A weekly route that looks perfect in the optimiser and blows up at 11:30am is usually failing for one reason: it was built around distance, not around the actual working day.

That matters most for weekly routes with split shift crews. If one crew starts early, breaks for lunch, then hands off to another crew, the software will happily stack stops in a way that looks efficient on paper and is impossible once you factor in travel back to base, lunch windows, and the person who inherits the route at midday.

The fix is not to “make the optimiser smarter” in the abstract. It is to model the day the way your operation really runs, then decide which parts of the route are allowed to move and which parts are not.

Build the route around work blocks, not a full-day fantasy

Split shift routing works best when you treat the morning and afternoon as two linked route blocks, not one continuous plan.

For weekly routes with split shift crews, that usually means defining:

• a morning block with its own start time, end time, and handoff point
• an afternoon block with its own capacity and depot return rules
• a fixed lunch or break window if the crew cannot stay on the road
• a handoff rule, such as “crew A finishes at the CBD, crew B starts from the same zone”

If you skip that structure, the optimiser will do what it is designed to do, squeeze every minute out of the route, often by pushing too many jobs into the first half of the shift and leaving the second half with slack. That is how overtime spikes happen. The first crew runs long, the second crew sits underused, and the handoff becomes a scramble.

A better setup is to cap each block by real labour hours, not just route duration. If the morning crew has 4.5 paid hours after breaks and the afternoon crew has 4.0, model those as hard limits. Then add a buffer, usually 10 to 15 per cent, for traffic, customer delays, loading, or a stop that takes longer than planned.

Key takeaway: If the route only works when every stop runs to the minute, it does not work.

The first week will expose bad service-time estimates

The first time you run recurring weekly stop routing on live data, the service times are usually wrong.

Not slightly wrong. Wrong enough to matter.

A stop marked as 12 minutes on the spreadsheet may take 6 minutes one week and 25 the next because of access issues, paperwork, pallet counts, key handovers, or a customer who insists on checking every item. In transport and distribution, that is normal. The optimiser does not know that unless you feed it real historical timings, and even then it needs a sensible range, not a single neat number.

What usually goes wrong first:

1. service times are copied from the booking system, not measured from the actual run
2. stop sequence is optimised for distance, then breaks when a site needs a specific arrival order
3. the first stop after lunch is too tight because the optimiser assumes the crew will restart instantly
4. handoff stops are treated like any other stop, so the plan keeps swapping them between crews

If you are setting up route planning for recurring delivery routes or service runs, start with a small sample of real weeks. Three to six weeks of actual run sheets is enough to show which stops are stable and which ones are time sinks.

For a related planning issue, see How do I optimise routes when my stops have strict time windows? Guide.

Use the data that changes the route, not the data that just looks tidy

Once you move past addresses and service times, the useful fields are the ones that affect whether a stop can actually be completed in that shift.

The fields that matter most are:

• stop duration, based on actual history
• preferred time window
• crew eligibility, if some crews handle certain sites better
• handoff eligibility, whether a stop can move between morning and afternoon
• site constraints, such as gate access, loading dock hours, or site contact availability
• recurring pattern, like weekly, fortnightly, or monthly
• exception status, such as “skip if stock not ready” or “call ahead required”

The fields that often look important but usually make routes worse are:

• too many priority flags
• customer notes that do not affect the drive or the service time
• overly granular scoring on “importance”
• postcode-based grouping when the actual barrier is site access or timing
• a long list of preferences that nobody maintains

That last one is the hidden maintenance burden. Every extra field becomes another thing that can drift out of date. A route plan with 40 fields and no ownership will decay faster than a simpler one with four or five fields that someone actually checks every week.

Stop trying to optimise the perfect week

Recurring route optimisation fails when it is built for a perfect week.

Perfect weeks do not happen. Someone calls in sick. A customer adds a standing stop. A truck is delayed at the depot. A site runs long because the pallet count changed. If the route only works when nothing changes, it is too brittle to trust.

The better approach is to build for variance. That means:

• keeping a buffer stop or buffer window in each shift
• marking one or two stops as movable rather than fixed
• allowing the optimiser to re-sequence within a zone, not across the whole week
• setting a maximum workload per crew so one long stop does not blow out the day

This is especially important for weekly routes with split shift crews, because a delay in the morning can cascade into the afternoon handoff. If the first crew runs 40 minutes late and the second crew inherits a route with no slack, the whole afternoon starts from behind.

If your operation also handles ad hoc service work, this is worth reading alongside How can I route service technicians so they arrive on time and still fit emergency calls in? Guide.

Lock the stops that need continuity, float the rest

A common mistake is locking everything because continuity feels safer.

It is not.

If every standing stop is fixed to one crew, you get dead mileage, uneven workloads, and more overtime on the side that inherited the longer run. If nothing is locked, the optimiser keeps shuffling stops between crews and you lose continuity with customers who expect the same person or the same sequence each week.

The right split is usually somewhere in the middle.

Lock these stops

• high-complexity sites where the crew knows the access process
• customers who expect the same technician or driver
• stops with strict site rules or repeat paperwork
• handoff-sensitive jobs that need the same sequence every week

Let these float

• short, low-risk stops
• flexible deliveries with wide windows
• stops near the handoff point
• jobs with predictable service times and no special access

A good rule is to lock the stops that create risk if they move, and float the stops that mainly create inefficiency if they stay fixed.

That is where recurring delivery routes become manageable instead of fragile.

Model exceptions without rebuilding the whole week

Weekly routes are mostly stable, which is why recurring route optimisation should never be rebuilt from scratch every Monday.

The least painful method is to build a base plan, then layer exceptions onto it.

Use one of these patterns:

• Base route plus exception list: the weekly core stays fixed, and only skipped or added stops are applied each cycle
• Standing stops with rotation rules: a stop can move between crews, but only within a defined zone or day
• Template by day type: Monday route, Tuesday route, and so on, with small edits instead of a full rebuild
• Freeze window: lock the route 24 to 48 hours before dispatch, then only allow emergency changes after that

That freeze window matters more than people think. If dispatch can keep tweaking the plan right up to departure, the crew never gets a stable run sheet and the optimiser never gets clean feedback on what actually happened.

For the operational side of that problem, How Do Route Optimisation Tools Handle Traffic? Recalculate is the right next read.

Why split shifts break naive optimisation

Split shifts are hard because the optimiser sees two separate problems and the business sees one day.

The software may think:

• morning crew has capacity left
• afternoon crew has capacity left
• therefore the route is fine

The operation sees:

• morning crew cannot take one more stop without missing lunch
• afternoon crew starts too far from the handoff point
• one customer wants the same person every week
• the route now needs a vehicle swap, a site call, and a paper handover

That is why weekly routes with split shift crews need a handoff rule that is explicit, not implied. Decide in advance whether the handoff is based on geography, stop type, time of day, or crew capability. Then keep that rule stable long enough to measure it.

If the handoff point keeps changing, the optimiser will keep reassigning the same stops back and forth because it is chasing the shortest path. That is not optimisation. That is churn.

A practical setup that actually holds up

If you are setting up route planning for recurring weekly stops, use this sequence.

1. Map the recurring stops by stability Separate fixed standing stops from flexible ones. Mark the stops that can move, the stops that cannot, and the stops that only move under exception.

2. Measure actual service times Use real run data, not booking estimates. If a stop has a wide spread, use the upper-middle of the range, not the average.

3. Split the day into route blocks Build morning and afternoon as separate capacity pools with a defined handoff.

4. Set hard constraints first Start times, break windows, maximum shift hours, and site access rules go in before optimisation.

5. Keep one buffer per shift Leave room for a long stop, a late truck, or a call-out.

6. Freeze the core route Lock the stops that create continuity or access risk. Float the rest within a zone.

7. Review exceptions weekly If the same exception appears three weeks in a row, it is not an exception. It is part of the route.

That last point is where most teams save the real time. A recurring route that keeps “unexpectedly” adding the same stop every Tuesday is not being disrupted by exceptions. It is revealing a standing pattern that should be modelled properly.

Where a logistics partner helps, and where it does not

If you have the route design nailed but the execution still falls apart because of storage, handover, freight handling, or cross-dock timing, the problem may be wider than route planning.

That is where end-to-end logistics solutions help, especially when warehousing, transport, and distribution all affect the same weekly run. In Australia, that matters in metro corridors where depot access, regional handovers, and interstate timing all collide in the same week. A provider like RAKC Group, with warehousing facilities in key logistics corridors and transport and distribution across metropolitan, regional, and interstate routes, can help keep the operational pieces aligned when the route itself is only part of the issue.

If your route is fine on paper but the handoffs, storage, or freight movement are creating the mess, look at the whole chain, not just the map.

The test that tells you whether the route is ready

Before you trust the optimiser, run this test on one week of real work:

• can the route survive one stop running 20 minutes long
• can a driver call-out be absorbed without rebuilding the full plan
• can a new standing stop be added without breaking the handoff
• can the afternoon crew start cleanly even if the morning crew finishes late
• can you explain, in plain language, why each locked stop is locked

If you cannot answer those cleanly, the route is still too rigid.

For teams building a broader control framework around this, How do I start a transport compliance program? Guide is the sensible companion piece.

The route you want is stable, not clever

The best recurring weekly stop routing is boring in the right way. The same jobs sit in the same places unless there is a real reason to move them. The split shift handoff is clear. The buffer is real. The exceptions are modelled instead of patched in at the last minute.

That is what keeps weekly routes with split shift crews from turning into a weekly rescue job.

If you want the faster path, RAKC Group can help build the transport side around the route, including warehousing, freight handling, and transport and distribution support that fits the way your weekly runs actually operate. Book a call and bring one week of run sheets, your standing stops, and the handoff rules you are using now. That is enough to see where the route is breaking.