How can I route service technicians so they arrive on time and still fit emergency calls in? Guide

The route that looks perfect at 7:00am is usually wrong by 10:15am

A technician route only works if it survives the first emergency call.

That is the real problem with commercial refrigeration emergency dispatch. You are not just planning stops, you are managing cold rooms, display cases, freezers, stock risk, and a dozen arrival promises that can turn into complaints if a tech is 20 minutes late. In Australia, where a failed refrigerated cabinet in a supermarket or a cool room in a hospitality venue can wipe out product fast, the schedule has to move without falling apart.

The fix is not to make every route “flexible”. That usually means vague windows, over-padded runs, and dispatchers quietly lying to themselves. The fix is to build a day plan that can absorb one emergency without forcing the whole board to restart.

Start with service class, not postcode

If every job is treated the same, the route will fail the moment a compressor fault comes in.

Split the day into three buckets:

• Fixed-window work, like planned maintenance or warranty visits with a committed arrival time
• Flexible work, like diagnostics, filter changes, or follow-up parts runs
• Interruptible work, which is your emergency pool for true refrigeration failures

That last bucket is where commercial refrigeration emergency dispatch lives. If you do not separate it from routine work, dispatchers end up stealing from Peter to pay Paul all day, and every ETA gets softer as the morning goes on.

This is also where field service routing needs to be honest about job type. A 20 minute thermostat replacement is not the same as a half-day leak hunt on a busy site with restricted access and no parts on hand. If your routing model does not distinguish those, it is not optimisation, it is guesswork.

Key takeaway: The route should protect the jobs with hard promises first, then use the remaining slack for emergencies, not the other way around.

The practical cutoff for same-day emergencies is not a time, it is a buffer

Most teams ask the wrong question. They ask, “Can we fit it in today?” The better question is, “Can we fit it in without breaking the next three commitments?”

A workable cutoff is usually based on three checks:

1. Travel time to the emergency site
2. Time left in the current job, plus realistic overrun
3. How many downstream jobs have fixed windows

If the emergency can be inserted only by delaying two or more fixed appointments, you are not dispatching, you are creating tomorrow’s complaint list.

For refrigeration work, the real pressure point is often the first 30 to 60 minutes after dispatch. Traffic changes, site access changes, and the technician may still be tied up on a job that looked like a 40 minute visit and has already become 90. That is why commercial refrigeration emergency dispatch needs a rule, not a hunch. A common rule is to hold a protected buffer in each zone, then release it only when the day is stable enough to absorb a call without pushing more than one non-emergency job outside its window.

What breaks first when the day shifts mid-job

The first thing that breaks is not the emergency call. It is the ETA chain behind it.

Say a tech finishes 25 minutes early in Parramatta, or 40 minutes late in Dandenong. The obvious move is to drop in the nearest job. That often causes a chain reaction because the next tech’s route was already built around a handover, a parts pickup, or a site access window. One early finish can create three late arrivals if the board is not recalculated from the actual current position.

The practical fix is to re-optimise from the live state, not from the morning plan.

Recalculate from these live inputs

• Technician GPS position and heading
• Actual job start and finish time
• Traffic now, not traffic at dispatch
• Parts on hand
• Site access constraints
• Job priority and SLA

If your route planning software cannot recalculate on those inputs, it will keep telling you a route is feasible when the field team knows it is not. That gap is common in service technician routing. The software sees a neat map. The technician sees a locked plant room, a customer who is still on lunch, and a job that needs a second pair of hands.

For a deeper look at live traffic behaviour and recalculation logic, see How Do Route Optimisation Tools Handle Traffic? Recalculate.

Trust the field, but calibrate the system with data

When dispatch software says a route is feasible and the tech says it is not, believe the tech first.

That does not mean the software is useless. It means the model is probably missing one of the things that actually eats time in refrigeration work, such as:

• Waiting for access cards or security sign-off
• Loading gas bottles, recovery gear, or replacement parts
• Parking at CBD sites
• Repeat visits because the fault needs diagnostics before parts can be confirmed
• Job overruns caused by contamination, icing, or electrical faults that look simple from the ticket

The way to calibrate the system is to compare planned time against actual time by job type, technician, region, and customer site. Do not use one average for everything. A 20 minute routine service in a suburban strip centre is not the same as a call into a multi-site account in inner Melbourne with loading dock restrictions.

A useful discipline is to review the last 20 to 50 jobs for each category and adjust the routing model to the median actual duration, not the best-case estimate. Then add a small, explicit buffer for the jobs that regularly run long. If the team keeps overrunning a certain customer or site type, encode it. Do not keep “hoping” the next visit will be different.

Stop padding every window, use protected slack instead

Dispatchers game the schedule for one reason. They have been burned too many times.

So they start padding every arrival window by 30 minutes, then 45, then a full hour. It feels safe. It also destroys customer trust, because now every ETA is vague and every tech looks late even when they are not.

The better approach is to protect slack in specific places:

• At the start of the day, for first-call delays
• In each zone, for travel spillover
• After known long jobs, for overrun risk
• In the emergency pool, for true refrigeration failures

That way you are not hiding time inside every appointment. You are reserving it where the risk actually sits. This matters in commercial refrigeration emergency dispatch because the emergency itself is unpredictable, but the pattern is not. If your same-day failures usually hit between 11:00am and 2:00pm, that is where the slack should live.

For teams that also manage freight movements or parts runs, the same principle applies to route discipline. If you are dealing with broader transport patterns as well, How Do I Manage Last-Mile Deliveries Without Route Chaos? is worth a look.

The best emergency cutoff is the one that protects the rest of the board

Dispatchers should say no when inserting the emergency would do more damage than the fault itself.

That sounds blunt, but it is the reality of commercial refrigeration emergency dispatch. A failed cool room at a hospitality venue in Sydney is serious. So is missing three planned maintenance visits that were already promised to different sites. The cutoff should be based on business impact, not whoever shouted loudest.

A simple decision rule helps:

Situation	Same-day insert?	Why
One flexible job can move 30 to 45 minutes	Yes	Low ripple effect
One fixed job moves, but no other ETAs change	Usually yes	Contained impact
Two or more fixed jobs move	Usually no	Chain reaction risk
Tech is already mid-urgent diagnostic	Only if another tech can take over	Avoids double disruption
Emergency is from a repeat high-risk account	Maybe, but reserve capacity for that account pattern	Prevents starvation elsewhere

The point is not to refuse emergencies. The point is to stop treating every emergency as equally urgent to the route. A freezer failure with stock loss risk is different from a non-critical check on a cabinet that is still holding temperature.

When the same few accounts keep calling, your model starts lying

This is where many teams get trapped. A handful of sites generate most of the emergencies, so the route model keeps bending around them. Routine maintenance gets pushed later and later until the “planned” work is really just whatever is left after the alarms stop ringing.

That is a bad pattern because it creates more emergencies. Missed maintenance on refrigeration assets leads to more breakdowns, which creates more emergency calls, which starves maintenance even further.

The workaround is to ring-fence capacity for those repeat accounts and separate them from the general emergency pool.

Do this instead

• Tag repeat emergency accounts by site and fault type
• Give them a standing response rule, not an ad hoc one
• Reserve a small daily or half-day capacity block for them
• Move routine maintenance into its own protected routing band
• Review whether the site needs planned service changes, not just faster response

If you are running regional coverage across Australia, this matters even more because a repeat call in one corridor can wipe out the routing efficiency for several nearby jobs. RAKC Group’s transport and distribution capability, with coverage across metropolitan, regional, and interstate routes, is the sort of structure that helps when you need to move people and parts without improvising every time. The value is not speed for its own sake. It is real-time visibility and communication so the board can be adjusted without guessing.

Route planning software is only as good as the rules you feed it

The software should not be deciding your policy. You should.

That means setting hard rules for:

• Maximum acceptable travel delay before a fixed job is protected
• Which job types can be pre-empted
• How much slack lives in each zone
• Which accounts get priority treatment
• When a route is frozen and when it can be recalculated

If you let the system optimise without those rules, it will happily produce a mathematically neat day that fails in the field. Good route planning software should support dynamic scheduling, not replace judgement.

For teams that need more than routing, end-to-end logistics support can help with the parts side as well, especially when a same-day repair depends on getting the right component to the right technician. RAKC Group’s Logistics Solutions and Tailored Business Support are relevant here because the real bottleneck is often not the drive time, it is the handoff between warehousing, transport, and the technician’s van.

A workable daily rhythm for dispatch teams

If you want the day to hold, use a rhythm the team can actually run.

Morning

• Lock fixed-window jobs first
• Assign flexible work into route clusters
• Hold emergency capacity in the busiest zone
• Check parts availability before releasing the first wave

Mid-morning

• Recalculate after the first two or three jobs finish
• Compare actual travel against plan
• Decide whether the emergency buffer is still intact

After lunch

• Use the reserved slack for true emergencies
• Move only one flexible job at a time if possible
• Do not keep stretching windows to “make it fit”

Late afternoon

• Protect end-of-day return travel and final appointments
• Stop accepting low-value inserts that will only create overtime and missed callbacks

That is the part many teams miss. Service technician routing is not a one-time planning exercise. It is a live control process. The route has to be corrected while the work is happening, not after the complaints arrive.

What good looks like on a hard day

On a bad day, the board still holds.

The tech who finished early gets the nearest flexible job. The emergency gets inserted only if it does not blow up the next fixed appointment. The team knows which accounts can displace routine work and which cannot. Dispatch can explain why a call was accepted or declined, and the answer is based on capacity, not panic.

That is what commercial refrigeration emergency dispatch should look like in practice. Not perfect timing. Controlled timing.

If you want the next step to be practical, start by reviewing the last 30 days of jobs and marking each one as fixed-window, flexible, or interruptible. Then compare planned versus actual duration by site type, not just by technician. You will see very quickly where your route is lying to you, and where your emergency buffer is too thin.

If you need help turning that into a working transport and distribution setup, RAKC Group’s Transport & Distribution and end-to-end logistics support are built for that kind of coordination, across metropolitan, regional, and interstate runs in Australia.