A serving-area stamp can look clean while one distribution lateral still sits on last week's cable ID, even after the feeder row already matched. Crews occupy a segment rather than a serving area. This gate is the occupancy call on that row, because a BOM line that still disagrees with the issued cable assignment is how a held lateral gets built.
This article starts after HLD and LLD already exist. We do not redefine those phases. Strand-count sizing stays on another worksheet, and splitter ratios stay with the PON sheet. Closure placement and general LLD QC are different reviews. Each HLD feeder and distribution segment ID, plus its approved count basis, has to join the current LLD artifacts a crew will actually use, on one issued revision, before occupancy.
What Feeder and Distribution Fiber Schedule Reconciliation Gates
Feeder and distribution fiber schedule reconciliation is Draftech's segment gate mapping each HLD feeder or distribution segment ID and its approved count basis to the current LLD cable assignment. The splice/interface schedule and BOM line sit on that same row with source revision and written delta. The output is 1 RELEASE or HOLD decision per segment.
The difference between HLD and LLD is already written. This gate does not retell it. It asks a narrower question: does the HLD object that left planning still join the three LLD artifacts a crew will actually use on the issued revision. If the join breaks, the segment stays on HOLD even when the rest of the serving area looks ready.
ITU-T L.250, Topologies for optical access network, approved 13 January 2024, is the edition in force. Clause 3.2.8 defines the feeder area as the area between the optical distribution frame and the distribution point. Clause 3.2.6 defines the distribution area as the area between the distribution point and the access point. Clause 3.2.7 places the distribution point where cables from access points gather and connect to the feeder cable from the central office. Those are topology names. They are not a release matrix.
The Fiber to the Home Councils Global Alliance Definitions of Terms, dated 8 March 2023, call the feeder network the fiber cables from the OLT in the central office to the first splitter in the field in a PON architecture. Homes Passed in that document excludes premises that still need substantial feeder and distribution cables. We keep those role names on the row so a feeder ID is not silently rewritten as a distribution ID at the cabinet when the count looks similar.
Draftech's in-house fiber network design writes this matrix into the issued set. Engineering stays in-house. When construction is in scope, Draftech-managed subcontract crews deliver full turnkey under our QA/QC and safety program. We do not claim a self-perform construction crew today.
RUS Bulletin 1753F-601a (PE-90a), RD-GD-2009-61, is the Minimum Performance Specification for Fiber Optic Cables (For Backbone, Feeder and Distribution Plant), with expiration 15 October 2065 on the bulletin. Scope is cable performance for the plant classes named in that title. It does not require this gate. If the project is not RUS work, that bulletin is not our default. We copy the plant-class names only when the contract names the bulletin.
Map Each HLD Segment ID to the LLD Cable
A title-block serving area is a weak join. Two laterals can share a street name and still be different HLD objects. The HLD segment ID is the planning key we refuse to drop. The LLD cable assignment is the sheath the mill tagged and the GIS record already uses. If those two keys do not name the same object, the row is not reconciled. It is a rumor with a stamp.
Approved count basis is the frozen input on that HLD object. How that number was sized is a different worksheet. Our job here is to confirm the LLD cable still carries that basis, not to invent a new one. A 144-count feeder on the HLD row that became a 96-count sheath on the LLD assignment is a delta, not a rounding note in the margin.
Role still matters on the same sheath. Our fiber backbone versus distribution design notes keep feeder and distribution as operating roles, with backbone kept on its own because a count does not rewrite the job. This matrix copies the role the HLD already assigned. It does not promote a distribution lateral into feeder because the count looks large.
Feeder and Distribution Fiber Schedule Reconciliation Matrix
| Matrix field | What it joins | HOLD if blank or mismatched | Authority on the row |
|---|---|---|---|
| HLD segment ID | The feeder or distribution object that left planning | Two laterals share a street label and the wrong one is occupied | Issued HLD segment register |
| Approved count basis | The frozen count already accepted for that object | The LLD sheath no longer carries that basis | Issued HLD count basis on that segment |
| LLD cable assignment | The current sheath ID on the issued LLD | The crew opens a cable the HLD never named | Issued LLD cable assignment |
| Splice/interface schedule | Where that sheath meets the next ID | A through path is treated as a tap, or the reverse | Issued splice/interface schedule |
| BOM line | The material line keyed to that same ID set | Procurement ships a count the schedule does not own | Issued BOM on the same revision |
| Source revision and delta | Which stamped set each artifact came from, and what changed | HLD Rev C is joined to LLD Rev A with no written delta | Revision block on the issued set |
| Owner, disposition, RELEASE or HOLD | Who may change the row, and whether the segment may be occupied | A serving-area stamp hides a held lateral | Named owner on this project control |
The table is a preview map. The next sections expand those rows. We do not treat the seven fields as a universal industry list. The operator may add columns. We do not drop these seven to make the sheet fit a title block.
This matrix is a Draftech project control, not an industry mandate. ITU-T L.250 names feeder and distribution topology. Broadband Forum TR-311 describes fiber infrastructure management architecture. Neither document requires a RELEASE or HOLD row. We run the gate because a serving-area stamp is how a held lateral gets occupied.
Join the Splice and Interface Schedule
A cable ID with no landing is not a path. The splice/interface schedule is the artifact that says where that sheath meets the next ID. FOA's Guide to Fiber Optic Network Design, in its Project Documentation chapter, already asks for the path of every fiber and the connections. Test records sit with that documentation. The FOA Standard for Installing Fiber Optic Cable Plants, 2025 V1, issued October 2025, Section 13.1, requires the path of every fiber to be identified through intermediate connections, including cable identification and splice locations. That is a path rule. It is not a placement lecture, and it is not an as-built splice book.
We copy the schedule ID onto the same row as the HLD segment ID and the LLD cable assignment. If the schedule still shows a through path and the LLD cable now terminates, the join is already broken. Hold it. If the schedule names a different cable ID than the assignment table, the join is broken in the other direction. Hold that too.
Interface here means the scheduled meeting of two IDs, not a new enclosure location. Where a closure can physically sit is a different sheet. This row only asks whether the named interface still belongs to this segment. A blank interface field is a HOLD, not a gap for occupancy day to fill.
One thing I still reject in our own reviews: a serving-area stamp that hides a distribution HOLD. A feeder row that already matches does not excuse a BOM line that does not. That limitation is on us. We now refuse a package that cannot name a decision on every feeder and distribution segment in the issued set.
That leftover serving-area stamp becomes a restoration problem at scale. The next artifacts still have to agree, or the path is only true on one sheet.
Join the BOM Line on the Same Revision
Procurement does not read the HLD narrative. It reads a line. The BOM line on this matrix is the material row keyed to the same HLD segment ID and the same LLD cable assignment. The splice/interface schedule has to match those keys. If the BOM still shows a 288-count reel after the assignment table moved to 144, the delta is already sitting in a warehouse waiting for a crew. Hold the segment until the issued revision carries one count.
The FTTH high-level design mistakes article already names the HLD-to-LLD handoff without a fiber-count reconciliation as a failure mode. This gate is the artifact that meeting was supposed to produce. It is not a second HLD review. It is the row that proves the approved count basis still owns a BOM line.
Quantity drift is the usual break. A route revision adds a lateral. The assignment table gets the new cable ID. The BOM keeps the old footage. We do not publish a universal tolerance for that gap. If the two artifacts disagree, the disposition is HOLD until the issued revision names the same footage and the same count basis. A percentage we invent is not engineering.
On rural electric cooperative plant, a feeder row that already matches is still not a serving-area release. Distribution laterals keep their own BOM lines. A mid-size ISP whose laterals sheet and BOM disagree is running two truths. The crew will occupy the one that arrived on the reel.
Record Revision, Delta, Owner, and Disposition
Four artifacts on four dates are not one design. Source revision is the stamped identity of each artifact on the row: HLD Rev C, LLD Rev B, schedule Rev B, BOM Rev A. Delta is the written change between those identities, in the units the artifacts already use. Owner is the named person who may change the row. Disposition is RELEASE or HOLD, written on that segment, not on the serving area.
Broadband Forum TR-311, Fiber Infrastructure Management System: Architecture and Requirements, Issue 1, approved 23 February 2015 and still in force as of this draft, describes a management layer with work-order management and fiber-network management plus a node-management function and resource data such as port and ID. That architecture can hold identifiers. It does not convert a GIS edit into an issued revision. If GIS shows the new cable ID and the stamped set still shows the old one, GIS is wrong until the revision is updated on the issued design.
The issued revision wins the mismatch. Editing GIS first is how a held distribution lateral becomes a working assignment with no design trail. Record the delta on the stamped set, then let inventory follow.
Owner is not a department name. A row that says "design" has no person to call when occupancy is already scheduled. We write a person, and we write what that person is allowed to change. Disposition is not a comment. RELEASE means this segment may be occupied under the issued revision. HOLD means it may not, and the written delta is the reason.
This gate is not the LLD quality control checklist. That page owns the 12-error review of the issued LLD package. A package can pass that review and still HOLD one lateral on this matrix. We do not rerun attachment heights or permit scales here. Slack-loop checks stay on that other page. We ask whether the HLD object still joins the three LLD artifacts on one revision.
Issue RELEASE or HOLD by Segment
Mixed results are the normal case when the package is otherwise ready. A feeder that already joins can move while an unmatched distribution lateral stays dark until the written delta is closed.
- RELEASE: the issued revision already names one object and one count, plus a named owner with occupancy permission for that segment.
- HOLD: the issued revision still disagrees with itself, or nobody is named to close the written delta.
Rural cooperative or single-state ISP under 5,000 passings: run this matrix before anyone occupies a feeder or a distribution lateral. Do not stamp the serving area. If the operator has no named owner yet, write one on the row before the first RELEASE. Hold GIS updates until the issued revision carries the same IDs.
Overbuilder on a 2026 BEAD schedule: lock the join at LLD, not at the HLD planning number. Grant closeout will ask what was assigned. A substitution from a 144-count sheath to a 96-count sheath that never hit this row is a records failure, even if the glass in the ground is sound. Segment-based HOLD lets one serving area stay dark while an adjacent feeder splices.
Campus or municipal telecom owner with dedicated circuits: keep PON take-rate math off these rows. Join by circuit ID from the equipment plan. A 12-count that carries eight working campus links still needs a BOM line and a schedule ID that match. A 24-count labeled "future" with no owner is a HOLD.
After a segment is released, construction readiness is a different gate. Our network design review for construction readiness compares the released design to current field state before NTP. This matrix does not replace that go/no-go. It is the occupancy permission that has to exist before that review has a released design to compare.
When four artifacts still carry four dates, Draftech's in-house fiber network design writes the HLD object, the LLD assignment, the splice schedule, and the BOM onto one issued revision with a named owner. A GIS edit is not a design change. We keep the matrix in-house so a department label is not the person a PM has to call when the BOM still shows 288 after the assignment moved to 144.
Active in 24 states. Available across all 50 U.S. states. If a live route needs this matrix before a segment is occupied, start with a free 20,000-ft design review. For a package already in draft, reach out at info@draftech.com.
Talk to our fiber design team about your segment matrix. Request a review of the occupancy matrix before the first held lateral is occupied.
