TBM and Heavy Civil Equipment KPIs: Utilization, Advance Rate, Weld Quality and Delivery

Tunnel boring and heavy civil equipment performance splits into two KPI families. Shop metrics belong to the fabricator: first-pass weld acceptance, bay utilization, plate yield and schedule adherence. Field metrics belong to the operator: machine utilization, advance rate, cutter change downtime and commissioning variance. Use the ranges below as reference points for spotting a gap worth investigating, not as targets to copy.

Every figure shows its basis. A superscript number links to a public source in the Sources list, and the sentence says what that source covers. A figure marked no cited source is kept exactly as originally published because no public source stating it has been found, so treat it as illustrative.

At a glance

Machine utilization and advance rate

Machine utilization, boring time divided by total available time, is the headline field KPI. Typical soft-ground earth pressure balance (EPB) projects run 30 to 40 percent no cited source, and world-class operations with disciplined ring logistics reach 50 to 60 percent no cited source.

For hard-rock machines, the Federal Highway Administration's 2009 road tunnel manual puts actual utilization in the range of 50 percent¹. Its example, two Norwegian tunnels, bored 51.5 percent of the time and lost the rest to regripping, cutter changes, machine and back-up downtime, muck haulage, services and other causes¹. Those hard-rock figures do not cover EPB drives.

Advance rate follows directly: 12 to 18 m per day is typical for a mid-size EPB no cited source, and 25 to 35 m per day is world class no cited source. Hard-rock gripper machines in favorable ground can exceed 40 m per day no cited source. For scale, the same manual notes that the disc cutter now lets tunnels advance tens or even hundreds of feet per day⁴.

Measure both from PLC time stamps, not shift reports, which flatter utilization by 5 to 8 points on most projects no cited source. Read weekly distributions rather than monthly averages.

For example, a 40 percent utilization month can hide a week at 12 percent that a weekly review would have caught in time to fix.

The biggest lever is rarely the machine itself; it is the supply chain behind it. Segment handling capacity should support ring builds of 25 to 45 minutes no cited source.

How to measure

• Divide boring time by total available time, using PLC time stamps.
• Record advance in meters per day.
• Review the weekly distribution, not only the monthly average.

What moves it

• Segment supply to the heading.
• Muck removal capacity.
• Cutter logistics.

Check that segment logistics can keep pace with the target advance using the Segment Handling Capacity calculator.

Cutter logistics

Cutter logistics carries three KPIs. Cutter consumption per 100 m of advance, trended against ground type, should stay within 20 percent of the geotechnical baseline no cited source; sustained excursions usually signal alignment or thrust problems rather than abrasive ground.

Downtime per cutter change runs 45 to 90 minutes typical no cited source and under 30 minutes world class in free air no cited source. Plan hyperbaric interventions as separate events.

Cutter stockouts should be zero, full stop. One stockout idles an operation burning $100,000 to $150,000 per day no cited source, so a 98 percent service level target for spare cutters is cheap insurance no cited source.

How to measure

• Count cutters changed per 100 m of advance, by ground type.
• Time each cutter change from stop to restart.
• Track spare cutter stock against the planned service level.

What moves it

• Alignment and thrust control.
• Planned cutter change windows, with hyperbaric interventions scheduled separately.
• Spare cutter buffers sized to the service level.

Trend cutter consumption against ground type with the Disc Cutter Wear Rate calculator.

First-pass weld acceptance and rework

On the fabrication side, first-pass acceptance at nondestructive testing (NDT) predicts everything else. TWI's 2011 industry survey found average repair rates of 1.0 percent of weld length for other welded structures and 2.0 percent for offshore structures, which means 98 to 99 percent of weld length passed first inspection on average².

World-class shops hold 98 percent or better no cited source, a level the survey averages above already reach. The survey reports averages and peaks across sectors, not a world-class tier, and it is not specific to tunnel equipment.

Repairs cluster, with peak rates up to 25 percent in specific locations such as root runs, fillet welds and limited-access areas⁵. Rework hours should stay under 3 percent of direct weld hours no cited source. Readings of 6 percent or more usually trace to two or three welders or one joint configuration no cited source, so Pareto by welder ID and joint type before spending anything.

How to measure

• Divide accepted weld meters by inspected weld meters.
• Use the planned inspection coverage as the denominator, so skipped joints cannot flatter the rate.
• Track rework hours as a share of direct weld hours, by welder ID and joint type.

What moves it

• Welder skill, which TWI's survey names with weld access and poor fit-up as the main perceived causes of repairs⁵.
• Access to the joint during welding.
• Fit-up accuracy before welding starts.

Set the inspection coverage that forms the denominator with the Weld Inspection Load calculator.

Steel plate yield

Steel plate yield belongs on the same board as weld quality. Typical is 70 to 75 percent no cited source, and world class is 82 to 88 percent with remnant tracking no cited source.

How to measure

• Divide net part weight by plate weight purchased.
• Report it per plate thickness and per job.

What moves it

• Remnant tracking and reuse.
• Nesting parts across jobs of the same plate thickness.

Check the yield of a nest before cutting with the Steel Plate Yield calculator.

Assembly bay utilization, schedule adherence and FAT

Assembly bay utilization is occupied hours with active work divided by available bay hours. It runs 60 to 70 percent in typical shops no cited source and 80 to 85 percent world class no cited source.

Above 90 percent, you have lost the surge capacity that single-piece work demands no cited source. Pair it with schedule adherence, milestones hit within 3 days of plan, where 70 percent is typical no cited source and above 90 percent is world class no cited source.

Factory acceptance test (FAT) first-pass rate belongs on the same board. World-class shops pass FAT on the first attempt 95 percent of the time no cited source, because hydraulic circuits are verified at the subassembly stage, not discovered failing at final test.

How to measure

• Divide occupied bay hours with active work by available bay hours.
• Count milestones hit within 3 days of plan.
• Record FAT results on the first attempt only.

What moves it

• Keeping some bay capacity free for surges.
• Milestone plans built on realistic durations.
• Verifying hydraulic circuits at the subassembly stage.

Verify hydraulic circuits against their load before final test with the Hydraulic Power Load calculator.

Delivery, commissioning and transport

Field and logistics KPIs close the loop. On-time delivery to site is 85 percent typical no cited source, and it only means something measured against the original contract date, never the last revised one.

For the top tier, the IndustryWeek Census of Manufacturers, a spring 2000 survey of U.S. plants, defined delivery leaders as plants with on-time delivery of at least 98 percent against the promised date³. That is cross-industry data, not specific to heavy equipment.

Field commissioning hours variance is the honesty metric. Typical shops overrun plan by 25 to 40 percent no cited source, while world-class shops land within 10 percent no cited source because they estimate bottom up and log hours daily.

Transport cost as a share of contract value should hold at 4 to 6 percent no cited source. Sustained readings above 8 percent mean split and routing decisions happen too late no cited source.

How to measure

• Measure on-time delivery against the original contract date.
• Compare actual commissioning hours with the plan, logged daily.
• Divide transport cost by contract value, per project.

What moves it

• Bottom-up commissioning estimates.
• Daily hour logging in the field.
• Making split and routing decisions at bid stage.

Price the largest fabrication at bid stage, before split and routing decisions are fixed, with the Cutterhead Fabrication Cost calculator.

Improvement discipline

Improvement discipline beats target setting. Review field KPIs weekly and shop KPIs at every milestone gate, and attack one constraint per quarter.

Teams chasing five KPIs at once typically move none, while shops that spent a quarter purely on first-pass weld acceptance report gains of 3 to 5 points that persist no cited source.

Rebaseline whenever conditions change, such as a new alloy, a new ground type or a new crane, because a benchmark carried over from the last project can be 30 percent wrong for this one no cited source. Publish the numbers where the crew works: posted utilization and acceptance boards are consistently worth 4 to 6 points on their own, before any process change is made no cited source.

How to build your own baseline

• Write down each KPI's definition and keep it fixed.
• Collect field data from PLC time stamps and shop data from inspection records.
• Compare against your own trend first and published ranges second.
• Rebaseline when the ground, the alloy or the equipment changes.