EV Charger BOM and AVL Localization

Prioritisation

Candidates are ranked on annual spend, lead time exposure, import dependence, qualification effort and certification risk. The first localization targets are high spend and low risk, which is usually mechanical, packaging and harness.

AVL structure

• Approved manufacturer part numbers, with the manufacturer named rather than the distributor
• Alternates listed explicitly, with the conditions under which each is valid
• Lifecycle status tracked, so obsolescence is seen coming
• Local supplier mapped against each approved part
• Validation status and the evidence supporting it

An approved vendor list that records only part numbers, without validation evidence, cannot answer the one question that matters during a field investigation: why was this part acceptable?

Alternate qualification

1. 1

   Datasheet comparison

   Parameter by parameter against the critical characteristics, not just the headline rating.

2. 2

   Sample inspection

   Physical and dimensional verification against specification.

3. 3

   Bench test

   Characterise the part itself under the conditions it will see.

4. 4

   Product integration

   Test in the actual product, because parts interact.

5. 5

   Environmental and EMC

   Where the part sits in those paths, verify it there.

6. 6

   Approval and AVL entry

   Add with recorded evidence and any conditions of use.

Supplier quality and change notification

Approved suppliers commit to notifying process, material and sub-supplier changes before they ship. Without that commitment, a qualified part can quietly become a different part between purchase orders.

Change control after approval

Approval is not permanent. Engineering change orders and notices govern any subsequent change, revisions are controlled, and the technical file stays aligned with what is actually being built.

Cost-down scorecard

Each candidate is scored on annual saving against qualification cost, test cost, certification risk and supply risk. Some savings are not worth the qualification effort, and the scorecard makes that visible before the work starts rather than after.

Firmware coupling in practice

• Meter protocol differences that change how energy is read
• Modem firmware and AT command variation affecting connectivity
• Flash memory timing and endurance differences
• MCU peripheral behaviour differences between silicon revisions
• Sensor characteristics that shift calibration

Each of these has caused field problems in EVSE programmes that qualified the part electrically and stopped there.

What the work covers

1. 1

   Local sourcing

   Working through the bill of materials to establish what can be bought in your market, what cannot, and what would need a design change to make it possible. The answer is rarely all or nothing.

2. 2

   Cost optimisation

   Taking cost out while the substitutions are being made, since the two decisions interact and doing them separately leaves money on the table.

3. 3

   Alternate qualification

   A second source for every part where a single supplier would stop the line. This is the item most often skipped and the one that halts production when it bites.

4. 4

   Obsolescence management

   Components go end of life over a product's decade. Planning for that during localisation is far cheaper than a redesign triggered by a supplier notice.

5. 5

   Documentation

   An approved vendor list and a controlled bill of materials the receiving factory can actually build to, rather than a spreadsheet that diverges within a year.

Why local sourcing is not only a cost decision

Cost is the usual reason to localise and it is not the only one. Sourcing in your own market shortens the supply chain, removes exchange rate exposure from the bill of materials, and reduces how much of your product depends on a border staying open.

For programmes where data residency matters, local manufacture also sits alongside local hosting: the units are built in your market and the platform they report to can run there too, so operational data stays within your own jurisdiction. Which of those reasons drives your programme changes what we optimise for, so it is worth saying at the start.

One category we treat differently

Internal cabling is a tempting line to value engineer and a bad one.

Cable quality and the way cable is routed inside the enclosure produce real field failures, and they pass every end of line test before appearing months later in a customer's car park. We found that in our own returns rather than in a standard. Where a substitution here saves money, we will usually say no and explain why, because the site visit costs more than the saving many times over.