The mechanical frame of a sortation system is usually the part that lasts. Loops, carriers and structural steel routinely run for fifteen years or more. The parts that age out first are the ones nobody sees: control processors, drive modules, input cards and the operating systems on the engineering workstations. When those go, the machine is mechanically fine and operationally stranded, and the repair becomes a project rather than a parts order.
Obsolescence is therefore a procurement question, not only a maintenance one. The decisions that determine how painful it will be are made at purchase, and they are cheap to make there. This article sets out what to ask for, how to classify the risk, and how to build a staged upgrade path that never requires a full shutdown.
Where Obsolescence Actually Bites
Three layers age at different rates, and the contract should treat them separately:
- Control hardware: processors, communication modules and input cards, typically the shortest commercial lifecycle in the system and the most disruptive to replace.
- Drives and motors: longer lived, but subject to manufacturer product retirement and to substitution that changes wiring, parameters and sometimes cabinet layout.
- Software and engineering tools: the quiet one. When the programming environment or the workstation operating system is unsupported, modifications become impossible long before the hardware fails.
That third layer is the one most often missed in a procurement review, and it is the one that turns a small change request into a migration project. Ask directly which engineering software version the machine is delivered against, what its support horizon is, and what happens when the supplier’s own toolset moves on.
What to Secure at Purchase
Four commitments convert obsolescence from a crisis into a scheduled cost:
- Component lifecycle statement: a written position on support horizon for the control platform, not a verbal assurance that it is a current model.
- Substitution policy: how the supplier handles a discontinued part, including who pays for the engineering when the replacement is not drop-in.
- Source files and documentation: the control source code, electrical drawings and parameter sets held in escrow or delivered, so a third party can support the machine if the supplier cannot.
- Spares for the discontinuation horizon: a last-time-buy position for the critical few items, held on the shelf.
The escrow point matters most. A sorter whose control source files exist only on the supplier’s engineering laptop is a machine you do not fully own, whatever the invoice says.
Building the Upgrade Path
The alternative to discovering obsolescence during a failure is to plan the migration in phases that each fit inside a maintenance window. The sequence most sites find workable starts at the edges: replace the engineering workstation and operating system first, then the human-machine interface and supervisory layer, then the communication network, and only then the control processors and drives. Each phase is independently testable, and a fault introduced in one does not take the whole line down.
Technical Background
Commercial components leaving production while installed equipment still works is the definition of obsolescence, and the industrial case usually arrives through the control platform rather than the mechanics. The hardware at the centre of it is the programmable logic controller, whose product generations retire faster than the machines they run. Systems kept in service after their support horizon become a legacy system, which is worth the description because it captures the real cost: not the hardware, but the shrinking pool of people and tools able to change it.
Whether an ageing line should be migrated or replaced is the question examined in the retrofit versus replacement decision guide, and the interface side of a controls migration is planned in WMS and PLC integration for sorters. The holding position for parts during the transition is set out in the critical spares list, with the contractual route to holding them covered in consignment and vendor-managed inventory. Support structures during and after migration are examined in maintenance planning and spare parts strategy, and the question of who owns the change is addressed in the OEM versus integrator procurement route.
Triggers That Should Start the Project
Do not wait for the failure. Four signals mean the migration project should already be scoped: the supplier issues an end-of-support notice for a platform you run; a routine spare arrives with a replacement part number and a wiring change; the engineering software will not install on a supported operating system; or the supplier’s own service team starts bringing a different toolset to site. Any one of these is manageable; all four together usually mean the upgrade has been left two years too late.
The discipline that prevents it is a short annual review. Once a year, ask which components in the machine are within two years of the end of their support horizon, what the substitution position is for each, and what the current phase of the migration plan is. Sites that do this spend on upgrades when it is convenient. Sites that do not spend the same money during an unplanned shutdown, plus the operating hours they lost getting there.
Frequently Asked Questions
How long do sorter control components last?
Varies by platform, but control processors and communication modules typically reach end of support well before the mechanical structure does, often inside ten years.
What is the most commonly missed obsolescence risk?
Engineering software and workstation operating systems. When they become unsupported, the machine becomes unmodifiable before any hardware has failed.
Should I require source code in escrow?
Yes where possible. Without the control source files and drawings, only the original supplier can support or modify the machine.
How should a controls upgrade be sequenced?
From the edges inward: workstation and interface first, then network, then processors and drives, so each phase is independently testable.
Is a full shutdown required?
Rarely. Phased migration is normally executed in maintenance windows, which is the main reason to plan it rather than react to it.
Who pays when a part is discontinued?
Whatever the substitution policy says, which is why it belongs in the purchase contract rather than in a service conversation later.
When should I consider replacement instead?
When several layers reach end of support together, or when the migrated machine still cannot meet the throughput and parcel profile the operation now needs.