Training volume measures coverage, not resolution

A completion count answers whether people attended, passed or retained a qualification at a point in time. It says little about which work they are authorised to perform, whether the right tools and information are available, how often the skill is practised or what happens when the documented procedure does not resolve the symptom.

A stronger capability map separates at least four things: safe reception and containment; standard diagnosis and repair; specialist intervention; and engineering-level exception resolution. Each needs a defined qualification, tool set, information access and decision boundary.

In 2022, Porsche described a high-voltage battery repair concept combining specialist bases, cross-regional hubs and mobile experts, with different qualification levels. Its now-archived account illustrates how specialist work and qualifications can be distributed across a network. It does not establish Porsche's current arrangements or the appropriate structure for another brand or market.

The exception begins with recognition and containment

A local workshop should solve the common cases locally. It should also know the exact point at which continuing becomes unsafe, wasteful or likely to destroy evidence.

For a high-voltage case, that may mean isolation state, battery classification, safe storage and transport. For software, it may mean preserving logs, version state and the sequence that produced the symptom. For ADAS, it may mean checking installation, wiring and preconditions before replacing a camera or control module. For an intermittent diagnostic case, it may mean capturing the right scan and environmental data before clearing codes or restarting the vehicle.

A September 2023 Stellantis STAR case, published through NHTSA, records front cameras and ADAS modules that were replaced and returned but found to have no internal failure. It directs technicians to inspect harness conditions and use scan details to distinguish repair paths. This is an ADAS diagnostic example, rather than evidence of an EV-specific failure pattern: the component named by a fault code may still require further investigation before replacement.

Local capability should be broad; scarce capability should be reachable

Routing every difficult vehicle to headquarters risks overloading scarce experts with work that a local site could perform with the right information or authority.

Local sites should own customer reception, safety, physical inspection, standard procedures, evidence capture, permitted repair and communication. Regional or central teams should hold capabilities that are scarce, product-wide or authority-heavy: deep diagnostic pattern recognition, software and log access, high-voltage specialist intervention, engineering liaison, exceptional warranty authority, constrained-part allocation and the decision to stop, transport or substitute the vehicle.

The boundary should follow case economics and risk. Replicate work that is frequent, time-critical and learnable. Concentrate work that is rare, safety-critical, tool-intensive or valuable because one team can see patterns across markets.

An escalation route is not yet an exception system

An email address or technical hotline gives the workshop somewhere to send a case. The service organisation still has to define what the receiving team can decide and how quickly it must respond.

The Stellantis note makes one boundary explicit: it offers diagnostic guidance but does not authorise warranty repairs. In a case requiring that approval, the technical answer and the authority to proceed therefore need to meet before the workshop can act.

A working exception system names the case owner, defines the minimum evidence pack, assigns severity, shows the current waiting state and sets a response clock for the next decision. The receiving expert needs authority: to request a test, authorise transport, release a constrained part, approve a repair path, involve engineering or decide that the customer needs an alternative vehicle.

It also needs a return path. A resolved exception should update a diagnostic tree, technical note, parts rule, training module or product feedback loop where appropriate. Otherwise the central team repeatedly solves the same case while the local network remains dependent.

Measure the wait between competent decisions

Vehicle-off-road time is an important outcome, but it is too late and too aggregated to govern the mechanism. The case should show where time is accumulating.

Useful states include waiting for initial triage, safe containment, diagnostic evidence, specialist review, repair authority, part, bay, customer decision and validation after repair. Each waiting state should identify the dependency preventing progress.

The proposed management measure is time to the next competent decision. Start the clock when progress requires information or authority the current case owner cannot supply, and stop it when a qualified person authorises a specific next action. A ticket acknowledgement or transfer does not stop the clock. Record the dependency and the person able to release it, so repeated delays point to the access, capacity or authority that needs changing.

Keep that decision clock separate from the time needed to carry out an authorised repair or deliver a part. Both contribute to vehicle-off-road time, but they require different management responses.

That measure prevents every delay being called a skills gap. If a trained technician waits three days for log access, the constraint is information. If diagnosis is complete but a battery cannot move, the constraint may be transport or safety process. If the repair is known but the part has no allocation owner, the constraint is supply authority.

Build from the exception log, not the training calendar

Start with recent difficult cases. For each one, reconstruct the first symptom, evidence available, decisions taken, hand-offs, waiting states, customer commitments and final resolution. Separate technical uncertainty from parts, capacity and warranty delay. Then group the exceptions by repeatability and risk.

The result should define what every local site must be able to do, what a smaller number of hubs must hold and what remains central. It should also expose the cases that should become standard work through better information, tooling or training.

Use recurring diagnostic gaps to set training priorities, and give access, capacity and authority failures to the owners who can change them. The objective is fewer vehicles waiting for someone who can make the next decision.