Backlog Level

Goal of the analysis:

Measure the level, age, and health of unfulfilled customer and production orders (“backlog”), understand why it is growing or shrinking, and determine actions to optimize throughput, on-time delivery (OTIF), and capacity utilization. Backlog Level quantifies open confirmed demand not yet shipped/completed, translates it into time at the constraint, and highlights risks (aging past promise, cancellations, mix). Executives use it to: (1) balance demand and supply in S&OP; (2) decide between overtime, debottlenecking, or lead-time repricing; (3) allocate scarce materials and capacity to priority customers/SKUs; and (4) improve scheduling, inventory positioning, and supplier performance.

Data required:

• Order book and fulfillment status:
  • Open sales/production orders by line: SKU, quantity, order date, customer priority/class, requested date, promise/commit date, partial-ship status, cancellations/returns.
  • Backlog flags (hold, credit, engineering change), expedite requests, substitution eligibility.
  • Bookings and shipments history to compute book-to-bill and backlog evolution.
• Capacity and routings:
  • Routings/standard hours per SKU and operation; bottleneck resource; effective capacity by line/shift (available hours, OEE/Availability × Performance × Quality).
  • Shift calendars, planned downtime (PM, sanitation), labor crew availability/skills.
• Inventory, materials, and WIP:
  • On-hand and on-order inventory by level (FG, WIP, critical components), ATP/CTP results, reservation/allocations.
  • Supplier POs, lead times, confirmations, delays, expedites; yield/quality holds impacting release.
• Scheduling and execution telemetry:
  • Finite schedule/dispatch list, actual start/finish times, changeover durations, downtime events (breakdowns, starved/blocked), scrap/rework.
• Customer and commercial context:
  • Service agreements (SLAs), penalty clauses, product/customer margins, allocation policies, lead-time quoting rules.
• Normalization and flags:
  • Seasonality, promotions, NPIs, engineering changes; M&A/new plant ramp; policy on what counts as backlog (frozen window vs outside, holds).

Detailed step-by-step instruction on how to conduct the analysis:

1. Define scope and metrics.
   • Backlog (units/$) = Σ open confirmed order quantity not shipped/closed.
   • Backlog Days (time at constraint) = (Σ Backlog standard hours at constraint) ÷ (Effective daily capacity at constraint).
   • Backlog Coverage = Backlog units ÷ Average daily shipments (by SKU family/plant).
   • Book-to-Bill = Bookings ÷ Shipments (ratio >1 grows backlog).
   • Aging buckets: 0–7, 8–14, 15–30, 31–60, >60 days from order date and from promise date; At-Risk Backlog = backlog aged > promised lead time.
   • Cancellation Rate = Cancelled backlog ÷ Opening backlog (per period).
2. Extract and cleanse data.
   • Pull open order lines, routings/standard times, capacity calendars, ATP/CTP snapshots, WIP and inventory positions.
   • De-duplicate split lines and partials; exclude unconfirmed quotes; tag holds/engineering changes; align time zones/calendars.
3. Convert backlog to capacity time.
   • Map each order line to its routing; multiply good units by standard hours at the bottleneck operation; adjust for expected yield/quality where relevant.
   • Compute Effective Capacity per day = Available hours × OEE (or Availability × Performance) for the constraint resource; include labor constraints if applicable.
   • Calculate Backlog Days and Coverage by SKU family, line, plant, and customer tier.
4. Profile and segment backlog.
   • Create aging distributions by SKU, customer tier, region, and promise class; highlight At-Risk Backlog (past promise).
   • Segment by material availability (ATP pass/fail), by operation (waiting for constraint vs upstream), and by substitution possibility.
   • Quantify margin-weighted backlog to guide allocation decisions.
5. Diagnose drivers of backlog build/burn.
   • Build a bridge: Opening Backlog → +Bookings → −Shipments → −Cancellations → +/− Scope changes (NPIs, engineering) → Closing Backlog.
   • Attribute shipment shortfalls to capacity (schedule adherence, changeover overruns, downtime), material shortages (supplier delay, quality holds), or mix (short runs, frequent setups).
   • Compute Backlog Burn Rate = Average weekly shipments applied to backlog (excluding new intake).
6. Link to schedule and capacity plans.
   • Overlay finite schedule and CRP: Required hours from backlog/plan vs available hours by week; flag overloads (>100%).
   • Evaluate feasibility of promised dates; re-commit using CTP with current capacity and material constraints.
7. Scenario and what-if analysis.
   • Simulate levers: SMED (−30–50% setup), overtime/extra shift, speed centerlining (+5–10% Performance), supplier expedite, SKU substitution, demand pushing/price actions.
   • Estimate time to clear backlog: Backlog Days ÷ (1 + uplift %) and project OTIF recovery curves.
8. Integrity checks.
   • Reconcile backlog units/$ to GL shipment accruals and inventory movements; ensure routings/standard times are current.
   • Confirm ATP/CTP logic aligns with policy (component alternates, yields); avoid double-counting WIP in both backlog and available-to-ship.

Format of the output of analysis:

• Executive scorecard: Backlog units/$, Backlog Days at constraint, Backlog Coverage, Book-to-Bill, At-Risk Backlog %, Cancellation Rate, and Backlog Burn Rate.
• Heatmaps: backlog by SKU family × plant/line, by customer tier, and by aging bucket; ATP/CTP pass/fail overlay.
• Bridges: backlog build/burn waterfall (bookings, shipments, cancellations), and driver attribution (capacity, materials, mix).
• Capacity view: weekly required hours vs available hours at constraint; projected backlog clearance date under current and scenario assumptions.
• Allocation panel: margin-weighted backlog, priority rules, and recommended re-commit dates for at-risk orders.

How to interpret results:

• High Backlog Days and rising Book-to-Bill (>1): Demand exceeds effective capacity; expect OTIF pressure and expediting. Prioritize debottlenecking and realistic re-promising.
• High backlog but low capacity utilization: Schedule adherence/material readiness issues rather than true capacity shortage; fix kitting, changeovers, and downtime.
• Backlog concentrated in few SKUs/customers: Targeted actions (supplier expedite, dedicated runs, substitutions) can materially reduce risk.
• Large At-Risk Backlog (past promise) or high cancellation rate: Service erosion; re-commit dates fast, allocate to strategic accounts, consider demand shaping/pricing.
• Very low backlog and low coverage: Underutilization risk; consider demand pull-forward, SKU consolidation onto constrained assets, or temporary line shutdowns.
• Volatile backlog with stable demand: Planning/schedule instability; tighten frozen windows and finite scheduling.

Steps a company can take to improve on this measure:

• Capacity and execution levers:
  • Increase effective capacity at the constraint: overtime/added shift, SMED to cut setup, centerline speeds, predictive maintenance to reduce breakdowns.
  • Improve schedule adherence: enforce frozen windows, stabilize material flow (kitting, supermarkets), and reduce resequencing/expedites.
• Supply and inventory levers:
  • Expedite/dual-source critical components; raise safety stocks temporarily for A SKUs; qualify alternates; pre-build subassemblies.
  • Deploy allocation rules and ATP with substitutions; prioritize high-margin/strategic customers.
• Commercial and demand shaping:
  • Adjust lead-time quotes to demonstrated capacity; offer incentives to shift demand (off-peak, alternative SKUs); implement order caps for long-tail/custom.
  • Reprice rush orders and low-margin SKUs that consume disproportionate setup time.
• Planning and governance:
  • Adopt finite capacity planning (CTP) integrated with current routings/OEE; daily backlog war room for at-risk orders; publish clearance plan with owners.
  • Link S&OP to demonstrated capacity; align procurement/PM shutdowns to demand valleys.
• Data and standards:
  • Refresh standard times and changeover matrices; automate backlog dashboards with aging/at-risk flags; ensure ATP/CTP accuracy.
• Example playbooks:
  • If Backlog Days at constraint = 18 with 25% setups, run SMED on top 3 recipe pairs (−40% setup) and add weekend shift for 4 weeks; target clearance to <7 days within 6–8 weeks.
  • If 35% of backlog is waiting on one component, qualify an alternate and re-sequence runs to consume available kits; cut at-risk backlog by ≥50% this month.
  • If cancellations exceed 6%/month, re-commit with realistic dates via CTP, add allocation to strategic customers, and introduce pricing for flexibility; aim <3% within a quarter.

Benchmark comparisons:

General benchmarks (directional):

• Make-to-Stock (stable demand): Backlog Coverage 3–10 days; At-Risk Backlog <5–10%; Book-to-Bill ~1.0 over a quarter.
• Make-to-Order/high mix: Backlog Days (constraint) 2–6 weeks depending on customization; At-Risk Backlog <10–15%.
• Engineer-to-Order/project-based: Backlog horizons in months; measure milestone adherence; cancellations typically <2–3% once released.
• Cancellation rate: Mature programs target <2–5%/month (sector dependent); spikes flag promise credibility or price issues.

Constructing internal benchmarks:

• Track 12–16 weeks of Backlog Days, Coverage, Book-to-Bill, At-Risk %, and cancellation rate by plant/SKU family/customer tier; publish quartiles and targets.
• Set guardrails (e.g., Backlog Days >10 for 3 weeks at constraint triggers war room; At-Risk >10% triggers re-commit/allocation actions).
• Pair backlog metrics with OTIF, schedule adherence, capacity utilization, and expedites to pinpoint systemic issues.
• Reset targets after major mix/portfolio changes, capacity upgrades, or supplier network shifts; maintain a backlog clearance playbook library.