Field Service Management and Service Parts
Scheduling software gets the technician to the site. Parts planning decides whether the right component is there when they arrive. Buyers frequently expect one product to do both and are disappointed, because they are different problems with different mathematics.
Scheduling and parts planning are different products. A suite that dispatches well may hold parts data without optimizing it. Establish which of the two problems you actually have.
Do not plan spare parts like finished goods. Failure-driven demand is intermittent, and deterministic methods systematically overstock slow movers while missing critical ones.
Availability is the objective, cost is the constraint. The question is where to place the next unit to buy the most service, not how to hold the least stock.
The recovery loop is a supply source. For repairable items, returns and refurbishment provide a large share of usable inventory and belong in the plan, not in a tail process.
Treat the improvement percentages as marketing. First-time fix and inventory reduction figures in this market come from vendors and are not independently verifiable.
Market overview
The short answer
Field service management software coordinates a mobile workforce: it manages work orders, schedules and dispatches technicians, runs the technician mobile application, checks warranty and entitlement, tracks contract and service level commitments, and increasingly ingests remote diagnostics. Service parts planning is a separate discipline that decides which parts to hold, in what quantity, and at which location, so that the component is present when the technician arrives. The two are frequently sold together and are not the same problem: scheduling is a routing and capacity problem, and parts planning is an intermittent-demand, multi-echelon inventory problem with a different objective function. The metric that joins them is the first-time fix rate, because a technician who arrives without the right part generates a second visit, and missing parts are the most common reason that happens.
KEY FACTS
Verified August 2026. Each statement below is complete on its own and cites its source in section 08.
What is in a field service management system?
The functional core is the work order and the person who executes it. Work order management captures what needs doing, against which asset, under which contract, and tracks it to completion. Scheduling and dispatch assign technicians to work, and this is where the software does its most computationally serious work: matching skills, parts, location, travel time, and service commitments across a workforce is a variant of a vehicle routing problem, and the operations research literature on technician routing and scheduling is substantial.
The technician mobile application is the delivery mechanism and the data capture point. It presents the job, the asset history, and the procedure, and it records what was done, what was consumed, and how long it took. That last capture matters beyond payroll: service time actuals are the input that makes future scheduling realistic, and deployments where technicians do not record reliably produce schedules that look plausible and cannot be executed.
Entitlement and warranty checking establishes whether the work is covered, by what, and who is billed. It is unglamorous and it protects margin directly, since unbilled work performed under an assumed warranty is a pure loss. Contract and service level management tracks the commitments that govern response and resolution times, which are the promises the scheduling engine is optimizing against. Remote and predictive diagnostics, where the equipment reports its own condition, increasingly initiate the work order before the customer calls, which changes the demand signal for both scheduling and parts.
What is generally not in the box is genuine parts optimization. Most field service suites hold parts data, show availability, and reserve stock against a job. Deciding how many of a slow-moving component to hold across a network of forward locations is a different computation, addressed in the next three sections, and buyers should establish explicitly whether a proposal includes it or merely displays the result.
Table 1. The capability split. The rows where both columns claim involvement are where proposals are most often ambiguous, and where a buyer should ask which system makes the decision rather than which one shows the result.
Why is service parts demand a different problem?
Finished goods demand is generated by customers deciding to buy. Service parts demand is generated by equipment failing, which is a different process with different statistics. An installed base of machines produces failures at rates governed by usage, age, and design, and for any individual part number those failures are sparse. The resulting demand pattern is intermittent: long stretches of zero, interrupted at unpredictable intervals by demand for one or a few units. Where the quantities also vary widely when they occur, the pattern is described as lumpy.
This breaks the methods that work on faster-moving items. Smoothing techniques designed for regular demand produce a small positive forecast in every period, which is wrong in the many periods with no demand and wrong again in the period when several units are needed at once. Percentage-based accuracy measures compound the problem by dividing by near-zero actuals. SCR covers intermittent-demand forecasting methods in its inventory optimization guide and does not repeat them here; the point for this page is that a service parts portfolio is dominated by exactly the items those methods exist for.
A second structural difference is criticality. In finished goods, a stockout costs a sale. In service, a stockout can idle a customer's production line, ground a vehicle, or breach a contractual response commitment with financial consequences. The cost of not having a part is therefore frequently unrelated to the value of the part itself, and inexpensive components can carry extremely high shortage costs. Planning that ranks items by value alone will under-protect precisely the cheap parts whose absence stops the fix.
Why is fill rate the objective rather than cost?
Service organizations sell availability. A maintenance contract promises that when equipment fails, it will be restored within a stated time, and the parts network exists to make that promise deliverable. The natural objective is therefore fill rate, meaning the share of demand satisfied immediately from stock at the location where it arose, measured against the service commitment rather than against a generic target.
Cost still matters, and it enters as a constraint rather than as the thing being minimized. The planning question is properly framed as: given this inventory budget, how should it be distributed across parts and locations to buy the most service. That is a different question from how to hold the least inventory, and it produces different answers, typically holding more of cheap critical items and fewer of expensive items whose absence is tolerable or coverable by an alternative.
Figure 1. The service loop and the recovery loop, with the first-time fix as the pivot between them. Parts availability at the point of service is what decides which branch the event takes, which is why the parts question and the scheduling question have to be answered together.
Two practical cautions follow. First, fill rate should be measured where the demand occurred, not at the central warehouse, since a part available centrally and absent at the technician's van has not served the event. Second, a single network-wide fill rate target applied uniformly is a weak instrument, because it forces the same protection onto items with wildly different criticality. Segmenting the portfolio and setting targets by segment is the standard remedy, and it is where most of the achievable improvement usually sits.
Table 2. Segmenting the portfolio. The third row is the one most often mishandled, because value-based ranking pushes cheap critical parts down the priority list even though their absence is what stops the repair.
How does multi-echelon positioning actually work?
Service networks are layered: a central warehouse, regional depots, forward stocking locations near customer sites, and frequently the technician's own vehicle. Positioning is the question of how many of each part to hold at each layer. Optimizing each location independently produces a predictable failure, because adjacent layers each buffer against the same uncertainty and the network as a whole holds more than it needs while still missing parts where they were required.
The foundational treatment is Sherbrooke's METRIC model, published in Operations Research in 1968, which addressed multi-echelon inventory for recoverable, meaning repairable, items and established how to evaluate a network as a system rather than as a set of independent locations. VARI-METRIC extended it to multiple indenture levels, where an assembly contains repairable subassemblies. Decades of subsequent peer-reviewed work relaxed the original assumptions, addressing limited repair capacity, lateral shipments between locations, and lost sales rather than backorders. A buyer does not need this literature to make a decision, and should know it exists, because it is what separates a product doing real optimization from one applying a rule per location.
The practical logic underneath is marginal analysis: where does the next unit of budget buy the most additional service. That framing is what produces the counterintuitive answers, such as holding a cheap part at every forward location while keeping an expensive one central even though it fails more often. The forward stocking location exists precisely to make this trade explicit: it trades holding cost for response time, and it is worth its cost only where the service commitment cannot be met from the layer behind it.
The data requirement should be stated plainly during evaluation. Multi-echelon optimization needs an accurate network definition, real replenishment and repair lead times rather than nominal ones, failure or consumption history at the location level, and a criticality classification. Organizations whose parts consumption is recorded centrally without location detail cannot support it, and the honest sequence there is to fix the data capture first.
What do the reverse flows and the service economics look like?
For repairable items the return flow is a supply source rather than an afterthought. A failed unit recovered from the field is inspected, repaired if economic, and returned to stock, which means the same physical population circulates between service, repair, and redeployment. Managing that population is the rotable pool problem: the pool must be large enough that a serviceable unit is available when one is needed, given how long repair takes and how many are in transit or on the bench at any moment. Repair turnaround time is therefore an inventory variable, and reducing it releases units in the same way that shortening a supplier lead time does.
This is where the boundary against SCR's returns guide sits. That page covers consumer returns, disposition economics, and the regulatory constraints on what may be done with returned goods. This page covers the industrial service loop, where the returned item is a component with a defined repair route and a known place in a circulating population. The two share vocabulary and are different operations.
On economics, service and aftermarket revenue frequently carries higher margin than the equipment it supports, and consultancy analysis has put the gap at a substantial multiple: one widely cited 2020 assessment stated that average global aftermarket operating margin ran at about two and a half times the margin on new equipment sales, and that aftermarket parts and services would continue to deliver more than half of a manufacturer's profit. These are consultancy estimates from firms that sell aftermarket advisory work, and no independent government or academic benchmark establishes them. The direction is corroborated by how manufacturers behave, and the specific multiples should be attributed rather than adopted.
Attach rate, meaning service revenue relative to product revenue, is the commercial metric that captures this, and first-time fix is the operational metric that protects it: repeat visits consume technician capacity that could have served new revenue, and they damage the customer relationship the contract renewal depends on.
The fair case against separating these disciplines deserves stating. Convergence is real: leading suites now bundle scheduling, parts visibility, and increasingly parts optimization, and connected diagnostics drive both the dispatch and the parts forecast from the same failure signal. A buyer could reasonably conclude that an integrated platform with weaker multi-echelon mathematics beats a best-of-breed pair, because the part, the technician, and the prediction live in one place and data continuity is what actually moves first-time fix. The boundary described here is a design choice rather than a law, and integration can win on exactly that argument.
Frequently asked questions
Is field service management the same as service parts planning?
No. Field service management schedules and dispatches technicians and manages work orders, entitlement, and mobile execution. Service parts planning decides which parts to hold, how many, and where. They integrate, and they are different products solving different mathematical problems.
What is included in a field service management suite?
Work order management, scheduling and dispatch, a technician mobile application, warranty and entitlement checking, contract and service level management, and increasingly remote or predictive diagnostics. Parts availability is usually displayed; parts optimization usually is not included.
Why can't I plan spare parts like finished goods?
Because demand comes from equipment failing rather than from customers ordering, which produces long runs of zero demand punctuated by unpredictable requirements. Methods built for regular demand produce a small positive forecast every period, which is wrong in both the quiet periods and the spike.
What is fill rate and why is it the objective?
Fill rate is the share of demand met immediately from stock at the location where it arose. It is the objective because service organizations sell availability against a response commitment, so cost enters as a constraint on how that availability is bought rather than as the thing being minimized.
What is a forward stocking location?
A small stocking point positioned close to customer sites, holding a limited range of parts so that a technician can collect what is needed without waiting for a central shipment. It trades holding cost for response time and is justified where the commitment cannot be met from the layer behind it.
What is multi-echelon optimization for parts?
Deciding stock levels across the whole network as one system rather than location by location, so that layers do not each buffer the same uncertainty. The foundational model is METRIC, published in 1968 for repairable items, later extended for multiple indenture levels.
What is the first-time fix rate?
The share of service events resolved on the initial visit. It matters because a repeat visit consumes technician capacity twice, delays the customer, and most commonly happens because a required part was not present, which is what links it to parts planning.
What are rotable pools?
A managed population of repairable units that circulates between service, repair, and redeployment. The pool must be sized against repair turnaround time and units in transit, which makes repair speed an inventory variable rather than only a cost.
Do service margins really exceed product margins?
Consultancy analysis says so, with one widely cited assessment putting aftermarket operating margin at roughly two and a half times new equipment margin. Those firms sell aftermarket advisory work, and no independent benchmark confirms the multiple, so treat the direction as credible and the number as attributed.
Where does field service software stop and inventory optimization begin?
At the decision. If the system displays what is available and reserves it against a job, it is field service software. If it decides how many units to hold at each location under an availability objective, it is doing parts planning, which is a different capability to buy and evaluate.
Method, sources, and where to go deeper
Method
The multi-echelon and repairable inventory treatment follows the primary peer-reviewed literature, principally Sherbrooke's 1968 METRIC paper and subsequent extensions, rather than vendor descriptions of those methods.
The scheduling discussion references the operations research literature on technician routing and scheduling rather than vendor claims about optimization quality.
Economic claims about aftermarket margin are attributed to the consultancies that produced them, with their commercial interest stated.
Supply Chain Research is independent and vendor-neutral. We accept no payment from the vendors or categories covered, and this page names no products.
Caveats
SCR publishes no benchmark for first-time fix improvement, inventory reduction, or service revenue uplift from software. The percentages circulating in this market come from field service and parts planning vendors and are single-client case studies without disclosed method.
Aftermarket margin multiples, including the frequently quoted figure of roughly two and a half times new equipment margin, originate with consultancies that sell aftermarket advisory services. They are estimates rather than audited or independently verified benchmarks.
The academic literature cited establishes how models behave under stated assumptions. It does not promise what a specific network will achieve, and model performance depends heavily on data quality.
Figure 1, Table 1, and Table 2 are structural summaries and decision aids rather than measured research findings.
Where to go deeper
Readers whose question is the mathematics of intermittent demand or multi-echelon optimization in general should read the SCR guide to inventory optimization and MEIO, which owns those methods and is deliberately not repeated here. The returns and reverse logistics guide covers consumer returns and disposition economics, distinct from the industrial repair loop described in section 06. The demand planning and forecasting guide covers forecasting method selection. The last mile guide covers vehicle routing, which shares mathematics with technician scheduling. Readers scoping across categories should start with the SCR supply chain software category map.
Sources
Sources
- Sherbrooke, C. C. METRIC: a multi-echelon technique for recoverable item control. Operations Research, 1968. Peer reviewed. The foundational multi-echelon model for repairable service parts.
- Springer. METRIC: a multi-echelon model, reference chapter. Academic reference covering the model and its extensions.
- Reliability Engineering and System Safety. Multi-echelon, multi-indenture spare parts inventory modeling. Peer reviewed. Multi-indenture extension of the multi-echelon problem.
- Soft Computing. Technician routing and scheduling problem, solution approach. Peer reviewed. Establishes the scheduling problem's structure and difficulty.
- Networks. Decision support for the technician routing and scheduling problem. Peer reviewed.
- arXiv. Extensions to the guaranteed service model for multi-echelon inventory optimization. Academic preprint on multi-echelon modeling assumptions.
- Deloitte Insights. Aftermarket services as a digital differentiator. Interested source: a consultancy that sells aftermarket advisory services. Source of the operating margin multiple cited in section 06.
- Boston Consulting Group. Aftermarket services and industrial manufacturer growth. Interested source: a consultancy. Cited for direction, not as a benchmark.
- PTC. Multi-echelon optimization for service parts management. Interested source: a software vendor. Cited only as an example of how the capability is described in the market.
