The Top-Cut — Ep. 2 show notes

Special — Benchmarking the Mine Plan

Both hosts are synthetic voices and the script is AI-generated. Every paper cited was verified against a primary source (DOI or arXiv ID) before publication, but the commentary is not that of practising researchers.

The Top-Cut — Show Notes

Special: Benchmarking the Mine Plan — Equipment Availability, Age, Automation and the Outside View

Traci and Joao apply the reference-class idea from Flyvbjerg and Gardner's How Big Things Get Done to the equipment numbers inside mine plans. They work through about forty sources: Luleå theses on loaders and drill rigs, CIM, AusIMM and SME papers going back to 1983, vendor handbooks and customer stories, open-pit fleet scheduling and dispatch studies, and four technical reports (Era Dorada, Las Chispas, Cariboo, Platreef). The result is a set of carry-forward rules of thumb. Each one is a benchmark range with its sources and the conditions under which it holds or breaks.

Every citation and every quoted number was checked against the source PDF, or for unread items against publisher metadata, before this episode was rendered. Page references below are to the PDF page unless marked "printed".

Both hosts are synthetic voices and the script is AI-generated. The commentary is not that of practising mining engineers, and the rules below are a starting reference class for a reviewer, not engineering advice.

The rules of thumb

Each rule gives the benchmark, then the sources behind it, then the conditions and how strong the evidence is.

1. Define availability before you argue about it

• Benchmark: None. This one is a precondition, not a number. Get it in writing: mechanical or operational availability; calendar or scheduled time; when downtime starts and ends; whether waiting for parts or fitters counts.
• Sources: In the Laval survey of 12 Canadian underground mines, almost half ended downtime when an operator took the machine from the garage, which mixes "available but idle" into downtime. Three logged downtime only by shift. Two could compute only mechanical availability. Five treated maintenance person-hours as downtime (Paraszczak, Lafontaine & Komljenovic 2008 / Lafontaine et al. 2007). At one mine in Paraszczak (2000), the downtime analysis was abandoned because of shift-level logging. LKAB did not record actual repair times (Gustafson 2013). At Boliden, about 80% of manually noted failure data was judged free of documentation errors (Al-Chalabi et al. 2014).
• Conditions: Applies everywhere. Most maintenance systems compute MTBF as working time divided by failures, which silently assumes a constant failure rate.

2. Productive time is about half of calendar time

• Benchmark: For underground mobile fleets, productive time is about 50% of calendar time, give or take 10 points. Drill percussion time is a few hours per rig per day.
• Sources:
• Telfer production drills: 83% available, 58% gross operating, 54% net operating, 59% utilisation of availability. Under 3.6 percussion hours per 12-h shift before optimisation (Adams et al. 2008).
• Boliden face rigs: 1.2 to 3.6 percussion h/day on one boom (Thunberg 2016).
• Finnish LHDs: effective use of 4–5 h per shift, and 5–6 h in the conclusion (Matikainen 1991).
• Young LHDs had about 66% utilisation; old ones under 50% (Paraszczak 2000).
• How the technical reports state it: Platreef 8.13 effective h of an 11.5-h shift; Cariboo 65–72% of a 10-h shift; Era Dorada FS 80% × 60% for drills and 75% × 70% for loaders and trucks.
• Conditions: Conventional, manned fleets. A plan implying much more than about 60% of calendar time as productive time needs site evidence. The evidence base is small, and mostly Scandinavian and Australian.

3. LHD availability falls with age

• Benchmark: 75–80% operational availability for a young fleet, sliding toward the mid-60s to low 70s past 15,000–20,000 h unless rebuilds or replacements are planned. Sustained 85%+ needs site evidence.
• Sources:
• About 89% uptime at ~3,500 engine hours versus 76% at ~14,800 h. Corrective maintenance went from 6% to 13% of scheduled time, and delays from 3% to 9% (Paraszczak 2000).
• About 73–75% for a Kiruna LHD in the late 1980s: MTBF ~51 h and MTTR ~17–18.5 h. The availability figure is our arithmetic on Kumar's printed data (Kumar 1989).
• LKAB's target was 9 of 13 LHDs available, about 69% (Gustafson 2013).
• Booysendal: 66% achieved against 85% planned on LHDs past 20,000 h, with a range of 53–81% (Malatji 2024).
• An 80% goal "can be achieved" (Matikainen 1991).
• Conditions: Assumes credible maintenance support. Preventive maintenance was only 13–25% of active maintenance in Paraszczak's mine and about 10% of maintenance work at LKAB. The evidence is old, from few sites, and includes one six-month MPhil sample.

4. Nameplate is not achieved

• Benchmark: Apply a job-efficiency haircut (the 50-minute hour, or whatever your site supports). Then apply a separate payload or fill haircut of 15–30% for blasted rock.
• Sources:
• Cat Handbook Ed. 50: all data assume "100% efficiency in operation" (PDF p.5). The excavator example uses 0.90 × 0.95 × 0.83 ≈ 71% (PDF p.70). Fill factors are 60–75% for well-blasted rock and 40–50% for poorly blasted rock (PDF p.55).
• Monitoring showed LHD buckets 20–30% below capacity (Matikainen 1991, citing Toro CECAM).
• Gol-e-Gohar trucks are nominally 105 t but carried 83–87 t (Dindarloo et al. 2015).
• Conditions: The two haircuts are separate. If payload-monitoring data exist, use them instead.

5. Maintenance cost rises in steps, not on a smooth curve

• Benchmark: Lifetime LHD maintenance is about 1–2.5× capital, with 1.5× as the central case over 12,000 h. Cost per hour steps up at component change-outs. A flat $/h over a long LoM is an inside-view tell. MMH/OH above about 0.3–0.4 is a replacement flag.
• Sources:
• 150% of capital over 12,000 h in good average conditions; 100% with excellent maintenance, up to 250% with poor (Hawcroft 1990, MINCOST'90).
• The maintenance-hours to engine-hours ratio went from 0.4 to 2 with age (Matikainen 1991, citing Toivanen 1989).
• MMH/OH of 0.17–0.22 for young LHDs and 0.42 for old ones, against a cited benchmark of 0.25–0.30 (Paraszczak 2000).
• Single-machine repair cost follows an "upward stairstep" (Cat Handbook, PDF p.353). Repair before failure costs about a third of repair after failure (PDF p.354).
• Truck $/h by age band is non-monotonic. The MIP schedule saved 21.6% against the spreadsheet "newest trucks first" approach (Fu et al. 2014).
• Actual LHD maintenance cost was about 2.4× plan (Malatji 2024).
• Conditions: The multiplier scales with life. Hawcroft himself says 150% is "totally inadequate" at a 22,100-h life.

6. Replacement and economic life

• Benchmark: UG drill rigs: about 7–10 years, which at Tara was ~23,000 operating h. LHDs: major overhaul or component change-out at about 10,000–14,000 h, with total life of about 20,000–25,000 h in older evidence. Haul trucks: about 100,000 h with an engine refurbishment at about 70,000 h (one case). The optimum is flat.
• Sources:
• One Boliden rig's data gave 96 months undiscounted (90–102), 115 months discounted (110–122), and 104 months with a redundant-rig cost (97–109), falling to 76 months at 6 cu/h. The reference group's estimate gave 96 months (Al-Chalabi et al. 2015a, 2015b, 2015c).
• Tara Simba: 81 months (79–84) at a 10% discount rate (Al-Chalabi 2022).
• Overhaul at 10–12k h, overhaul once, maximum 20–25k h (Matikainen 1991).
• Operators' lives ranged from 10,000 to 22,100 h (Hawcroft 1990).
• LKAB's supplier plan changed engine, converter and gearbox at 13–14k h, the transmission at 10k and the hydraulic pump at 8k (Gustafson 2013).
• The R1700 rear engine mounts target 12–14k h to align with the C13 engine's planned component replacement schedule. This is mount life, not engine life (Cat blog 2024).
• 100,000 h truck life with engine refurbishment at 70,000 h (Fu et al. 2014).
• Conditions: Boliden's cost data before month 76 were not recorded and had to be extrapolated. Tara's fuel use and metres drilled were random numbers within expert ranges. The method matters: a published heuristic came out about 10% worse than the IP optimum (Sherer & Gentry 1983). Any LoM over about 10 years without a timed fleet renewal has a hole in it.

7. MTBF is fine for availability, dangerous for maintenance planning

• Benchmark: A whole LHD fails every ~30–50 operating hours. Face drill rigs have a characteristic time between failures under a shift. Hoses, feeders and rock drills account for ~63% of failures.
• Sources:
• Paraszczak (2000): MTBF of 31.7 and 36.2 engine hours, with the same subsystem ranking at both mines.
• Kumar (1989): MTBF of ~51 h. For the engine, exponential and Weibull MTBF agree within 0.5%, but 100-h reliability is 86% (Weibull) against 60% (exponential), about a 30% error.
• Kumar (1990): fitting iid models to deteriorating hydraulic systems wrongly showed a constant or decreasing failure rate.
• Hoseinie et al. (2018): Weibull scale of 7–11 h.
• Conditions: Ask for a trend test (Laplace or TTT plot) before any ageing or interval analysis.

8. Automation buys hours, not speed

• Benchmark: About 5–15% more output from working through shift change, blast re-entry and travel. Ramp it in. Don't credit higher instantaneous productivity.
• Sources:
• LKAB: the semi-automatic LHD moved about 13% fewer t per engine hour (273 vs 315) for the same t/bucket and the same maintenance cost per tonne. Its transmission was less reliable. 75% of 1,411 idle stops were mining-related; only 3.5% were the automation system (Gustafson 2013).
• Telfer: manual holes cycled 19% faster. First-hole automation reliability rose from 39% to over 77% in about 8 months. Of the +14.8% fleet gain, about 10.7 points came from standardising drilling practice, and the automated between-shift share rose from 2.5% to 6.6% (Adams et al. 2008).
• Vendor items (mechanisms only):
• Zinkgruvan: +4 drilling h/day, but "still considered to be a test".
• Éléonore: ~9% of metres drilled tele-remote; the manager cites 5–10% more footage.
• Sasa: +50%, with no baseline.
• DeGrussa jumbo: +15% gross penetration rate, while between-shift drilling was not yet validated (Välivaara 2016).
• Conditions: Size the benefit against mining-related disturbances. The 1991 promise of +40–60–(100)% effective time (Matikainen) is still a promise.

9. Fleet sizing: match first, then dispatch, and calibrate simulations

• Benchmark: Get the match factor close to 1. Dispatch is worth single digits up to about 20%. A simulation that hasn't reproduced the mine's actual production doesn't count.
• Sources:
• Shovel-limited plateau: 1 shovel capped at 5.42 Mt/a, and truck utilisation fell from 96% to 60% (Torkamani & Askari-Nasab 2012).
• With 2 shovels, output maxed at 22 trucks; 3 and 4 shovels gave about the same (Dindarloo et al. 2015).
• Sungun: match factor of 0.54–0.80. Fixing it gave +40% in simulation; dispatch added +7.8% (Saadatmand Hashemi & Sattarvand 2015).
• Field reports cite 7–20% for DISPATCH at 13 mines (review in Nehring et al. 2010b).
• A Kenyan quarry ran 16 trucks where 12 was optimal (Elijah et al. 2021).
• Actual production was 55–82% of plan, while a global-dispatch simulation exceeded actual by 90–270% (Chaowasakoo et al. 2017).
• Conditions: Most numbers here are simulated. Check that breakdowns and mining disturbances are in the model.

10. MIP schedules are worth single digits of NPV

• Benchmark: A credible uplift over a competent manual schedule is single-digit NPV. Double-digit claims come from conceptual or unimplemented cases.
• Sources:
• Mount Isa: +23%, not implemented (Trout 1995, via Little et al. 2008).
• Conceptual 9-stope case: +0.66%. Conceptual 50-stope case: +5.44% (Nehring et al. 2010a).
• Kiruna: deviation from target cut from 10–20% to 6%, and implemented (Topal, via Little et al. 2008).
• Conditions: Deterministic schedules exclude breakdowns, so apply the same haircuts as everything else.

Test cases: the four technical reports

• Era Dorada, S-K 1300 TRS Initial Assessment (GE21 for Aura; June 2025). §13.8 (printed p.166) and §13.9 (printed p.167).
• What it states: 2-boom jumbos at "two to three rounds per shift"; 4.4 m advance per round; a flat fleet table for Y1–Y17.
• What's missing: No availability, utilisation, productivity rates or replacement lives are stated, so these can't be tested. §13.11 says the rate "applied indexes considered appropriate", and §18 says "manufacturer database equipment usage rates" were used.
• Flags: The text describes a "6.0 yd³ (10 t)" LHD while the table lists 10 t at 4.0 m³. Sustaining capital of US$136M includes fleet renewal, but no timing is given.
• Era Dorada, S-K 1300 TRS & Feasibility Study (Ausenco / Snowden Optiro; Dec 2025). §13.5 Table 13-8 (printed p.222).
• What it states: Drills 80% availability and 60% use of availability (48% of calendar). LHDs and trucks 75% and 70% (52.5%). LHD 10 t at 86 t/h, 7 t at 59 t/h; 30 t truck averaging 39 t/h.
• Verdict: Inside the reference class, close to Telfer's 83% / 59% / 54%.
• Flags: 75% LHD availability is a young-fleet number over an 18-year LoM. The fleet-sizing Tables 13-9 and 13-10 are referenced but absent from the copy reviewed. The "150 m/jumbo.month" annotation is ambiguous (if it is advance, the IA's rounds-per-shift rate is several times faster).
• Las Chispas, NI 43-101 FS (Ausenco / GMS; effective 4 Jan 2021).
• What it does well: Production rate set from empirical rules (Taylor ~1,000, Long ~1,300, McIsaac & Devon Smith ~1,400 t/d; 1,250 chosen). Single-heading advance of 2.5 m/d, capped at 300 m/month, "shown to be achievable at Las Chispas in 2019–2020". Both are outside-view practice.
• What's missing: Contractor fleet availability and productivity are not disclosed. The ventilation utilisation (80% / 50%) is "conservative" for air, which runs in the opposite direction to productivity.
• Cariboo Gold, NI 43-101 FS (Osisko Development / InnovExplo; Jan 2023).
• What it states: Rates "based on experienced mining contractor feedback or typical rates for similar operations". Effective productivity of 72% (development) and 65% (manual LHD and trucks) of a 10-h shift. LHD stope mucking at 887 t/d manual, 969 tele-remote between shifts (+9%), 1,027 automated (+16%). 515 of 4,900 t/d come from autonomous operation over the shift change, "assuming optimal setup conditions".
• Good practice: An independent rebuild schedule.
• Verdict: Automation credit has the right mechanism and is at the top of the range. Ask for a ramp-up.
• Platreef, 2022 FS (Ivanhoe; March 2022).
• What it states: 8.13 effective h of an 11.5-h shift (71%). Swing units added "to account for overall availability", with no % given. Single heading at 4.3–5.8 m/d from first principles.
• Flags: No change to activity durations, development or stoping rates since the 2017 FS. The study itself recommends further utilisation analysis "to reduce fleet size".

Four reports are test cases, not a sample. Across them, single-heading rates run from 2.5 m/d (demonstrated) to 4.3–5.8 m/d (first principles).

Bibliography

Framework

• Flyvbjerg, B. & Gardner, D. (2023) How Big Things Get Done. New York: Currency (Crown). ISBN 978-0-593-23951-3. The source of the inside-view / outside-view and reference-class forecasting framing. We use the framework only, not the book's project statistics.

Availability, reliability and data quality

• Paraszczak, J. (2000) "Failure and downtime reporting — a key to improve mine equipment performance." CIM Bulletin 93(1044):73–77.
• Paraszczak, J., Lafontaine, E. & Komljenovic, D. (2008) "Follow-up of production equipment availability and its contributing factors in selected Canadian underground mines." CIM Bulletin 101(1107), Paper 4. Conference version: Lafontaine, E., Paraszczak, J. & Komljenovic, D. (2007), Montreal Energy & Mines, 29 Apr–2 May 2007, pp.119ff. Cited as one work.
• Kumar, U. (1989) "Availability studies of load-haul-dump machines." Proc. 21st APCOM, SME-AIME, Las Vegas, 27 Feb–2 Mar 1989, pp.323–335. Reprinted as Paper III in Kumar (1990).
• Kumar, U. (1990) Reliability Analysis of Load-Haul-Dump Machines. Doctoral thesis 1990:88 D, Luleå University of Technology.
• Kumar, U. & Klefsjö, B. (1992) "Reliability analysis of hydraulic systems of LHD machines using the power law process model." Reliability Engineering & System Safety 35(3):217–224. DOI: 10.1016/0951-8320(92)90080-590080-5). Read in its pre-publication version (Paper V in Kumar 1990).
• Al-Chalabi, H. (2015) Reliability and Life Cycle Cost Modelling of Mining Drilling Rigs. PhD thesis, Luleå University of Technology (DiVA diva2:990356).
• Al-Chalabi, H., Lundberg, J., Wijaya, A. & Ghodrati, B. (2014) "Downtime analysis of drilling machines and suggestions for improvements." Journal of Quality in Maintenance Engineering 20(4):306–332. DOI: 10.1108/JQME-11-2012-0038. Paper I in the thesis.
• Hoseinie, S.H., Al-Chalabi, H. & Ghodrati, B. (2018) "Comparison between simulation and analytical methods in reliability data analysis: a case study on face drilling rigs." Data 3(2):12. DOI: 10.3390/data3020012.
• Thunberg, H. (2016) Reliability-based Maintenance Plan for Underground Drilling Rigs. MSc thesis, Luleå University of Technology.
• Dindarloo, S.R. (2016) "Support vector machine regression analysis of LHD failures." International Journal of Mining, Reclamation and Environment 30(1):64–69 (online 2014). DOI: 10.1080/17480930.2014.973637.

Maintenance cost, economic life and replacement

• Sherer, H.E. & Gentry, D.W. (1983) "Practical aspects of mine equipment replacement analysis." Mining Engineering (November 1983):1552–1555. From SME preprint 82-311, SME-AIME Fall Meeting, Honolulu, Sept 1982 (16-pp preprint also read). Filed in the library as "SME 1984". The journal issue is November 1983.
• Hawcroft, D.P. (1990) "Hourly ownership and operating cost projection — rubber tyred loaders." MINCOST'90, AusIMM, Sydney, June 1990, pp.77–80.
• Matikainen, R. (1991) "How to increase LHD productivity in hard rock mining." Reliability, Production and Control in Coal Mines, AusIMM, Wollongong, 2–6 Sept 1991, pp.85–88.
• Al-Chalabi, H., Lundberg, J., Ahmadi, A. & Jonsson, A. (2015a) "Case study: model for economic lifetime of drilling machines in the Swedish mining industry." The Engineering Economist 60(2):138–154. DOI: 10.1080/0013791X.2014.952466. Paper II in the thesis.
• Al-Chalabi, H., Lundberg, J. & Jonsson, A. (2015b) "Economic lifetime of a drilling machine: a case study on mining industry." International Journal of Strategic Engineering Asset Management 2(2):177–189. DOI: 10.1504/IJSEAM.2015.070624.
• Al-Chalabi, H., Lundberg, J., Al-Gburi, M., Ahmadi, A. & Ghodrati, B. (2015c) "Model for economic replacement time of mining production rigs including redundant rig costs." Journal of Quality in Maintenance Engineering 21(2):207–226. DOI: 10.1108/JQME-07-2014-0041. Read in the thesis version (Paper IV).
• Al-Chalabi, H., Ahmadzadeh, F., Lundberg, J. & Ghodrati, B. (2014) "Economic lifetime prediction of a mining drilling machine using an artificial neural network." International Journal of Mining, Reclamation and Environment 28(5):311–322. DOI: 10.1080/17480930.2014.942519. Paper III in the thesis.
• Al-Chalabi, H. (2022) "Development of an economic replacement time model for mining equipment: a case study." Life Cycle Reliability and Safety Engineering 11(2):203–217. DOI: 10.1007/s41872-022-00188-1.
• Malatji, P. (2024) Assessment of factors affecting maintenance costs of load, haul, and dump machines at a South African platinum mine. MPhil (Engineering Management) minor dissertation, University of Johannesburg.
• Fu, Z., Topal, E. & Erten, O. (2014) "Optimisation of a mixed truck fleet schedule through a mathematical model considering a new truck-purchase option." Mining Technology 123(1):30–35. DOI: 10.1179/1743286314Y.0000000055.

Productivity and automation

• Adams, H.M., Boniwell, P. & Knights, P.F. (2008) "Production drill optimisation at Telfer Underground — a move towards automation." Tenth Underground Operators' Conference, AusIMM, Launceston, 14–16 April 2008, pp.37–40.
• Välivaara, J. (2016) "Automated drilling features for improving productivity and reducing costs in underground development." MassMin 2016, Sydney, 9–11 May 2016, pp.723–730. The author is a Sandvik product manager.
• Gustafson, A. (2013) Automation of Load Haul Dump Machines: Comparative Performance Analysis and Maintenance Modeling. Doctoral thesis, Luleå University of Technology. ISBN 978-91-7439-760-4.
• Gustafson, A., Schunnesson, H., Galar, D. & Kumar, U. (2013) "Production and maintenance performance analysis: manual versus semi-automatic LHDs." Journal of Quality in Maintenance Engineering 19(1):74–88. DOI: 10.1108/13552511311304492. Paper I in Gustafson (2013).
• Gustafson, A., Schunnesson, H., Galar, D. & Kumar, U. (2013) "The influence of the operating environment on manual and automated load-haul-dump machines: a fault tree analysis." International Journal of Mining, Reclamation and Environment 27(2):75–87. DOI: 10.1080/1755182X.2011.651371. Paper II.
• Gustafson, A., Schunnesson, H. & Kumar, U. (2015) "Reliability analysis and comparison between automatic and manual load haul dump machines." Quality and Reliability Engineering International 31(3):523–531. DOI: 10.1002/qre.1610. Paper III. It is also the data source for Dindarloo (2016).
• Siva Prasad, B.N.V., Murthy, V.M.S.R. & Pandey, S.K. (2016) "Investigations on rock drillability applied to underground mine development vis-à-vis drill selection." Recent Advances in Rock Engineering (RARE 2016), Atlantis Press, pp.399–403. DOI: 10.2991/rare-16.2016.62. Substitute for the 2022 chapter below. Not discussed on air. Note that its "per month" and "per annum" figures are internally inconsistent.

Scheduling, dispatch and fleet sizing

• Little, J., Nehring, M. & Topal, E. (2008) "A new mixed-integer programming model for mine production scheduling optimisation in sublevel stope mining." Australian Mining Technology Conference, AusIMM, 16–18 Sept 2008, pp.157–172. Its literature review is the source for Trout (1995) and Topal (2003).
• Nehring, M., Topal, E. & Little, J. (2010a) "A new mathematical programming model for production schedule optimization in underground mining operations." Journal of the Southern African Institute of Mining and Metallurgy 110(8):437–446.
• Nehring, M., Topal, E. & Knights, P. (2010b) "Dynamic short term production scheduling and machine allocation in underground mining using mathematical programming." Mining Technology 119(4):212–220. DOI: 10.1179/1743286310Y.0000000001.
• Torkamani, E. & Askari-Nasab, H. (2012) "Verifying short-term production schedules using truck-shovel simulation." In Askari-Nasab, H. (ed.), Mining Optimization Laboratory Report Four, University of Alberta, ISBN 978-1-55195-301-4, pp.190–205.
• Dindarloo, S.R., Osanloo, M. & Frimpong, S. (2015) "A stochastic simulation framework for truck and shovel selection and sizing in open pit mines." Journal of the Southern African Institute of Mining and Metallurgy 115(3):209–219.
• Saadatmand Hashemi, A. & Sattarvand, J. (2015) "Simulation based investigation of different fleet management paradigms in open pit mines — a case study of Sungun copper mine." Archives of Mining Sciences 60(1):195–208. DOI: 10.1515/amsc-2015-0013.
• Chaowasakoo, P., Seppälä, H., Koivo, H. & Zhou, Q. (2017) "Digitalization of mine operations: scenarios to benefit in real-time truck dispatching." International Journal of Mining Science and Technology 27(2):229–236. DOI: 10.1016/j.ijmst.2017.01.007.
• Elijah, K., Joseph, G., Samuel, M. & Mauti, D. (2021) "Optimisation of shovel-truck haulage system in an open pit using queuing approach." Arabian Journal of Geosciences 14:973. DOI: 10.1007/s12517-021-07365-z.
• Bastos, G.S. (2010) Methods for Truck Dispatching in Open-Pit Mining. DSc thesis, Instituto Tecnológico de Aeronáutica, São José dos Campos. Read for background: uncertainty is absent from most commercial dispatch systems. Its test is on a simulated example mine only.

Vendor and trade press (treated as the inside view)

• Caterpillar (2022) Caterpillar Performance Handbook, Edition 50 (June 2022). Introduction (PDF p.5); excavator production example and job-efficiency table (PDF pp.70, 74); bucket fill factors (PDF p.55); Owning & Operating Costs, §22, repair costs (PDF pp.353–354).
• Caterpillar (2024) "The Cat R1700 Underground Loader — Continuous Improvement." Cat Mining blog, 23 May 2024. The 12,000–14,000 h figure is the target life of the redesigned rear engine mounts, set to align with the C13 engine's planned component replacement schedule. It is not engine life.
• Epiroc / Atlas Copco (2015) "Milestone for Boomer S1 D as 500th rig goes to Macedonia." Customer story, 11 Dec 2015 (Sasa mine).
• Epiroc (2018) Mining & Construction Canada Vol. 2, 2018: Red Chris Pit Viper 351 rebuild; Éléonore tele-remote Simba.
• Epiroc (2022) "Zinkgruvan Mining's big step forward — Teleremote via dedicated network." Customer story, 20 Jan 2022.
• Epiroc (2025) "A success story: transforming underground drilling with the Epiroc Simba E60 S." Customer story (Boliden Somincor).
• Mining.com / Epiroc (2018) "Epiroc COP MD 20 rocks the drilling industry." 16 April 2018. Vendor claims: a 1,000-h service interval, 67% longer than the COP 1838HD+, and an estimated 25% lower maintenance cost. Not discussed on air.

Technical reports (test cases)

• GE21 Consultoria Mineral (2025) S-K 1300 Technical Report Summary, Initial Assessment, Era Dorada Gold Project, Jutiapa, Guatemala, for Aura Minerals Inc. Effective 31 Dec 2024, issued 2 June 2025. §13.8–13.9, printed pp.166–167.
• Ausenco do Brasil et al. (2025) Era Dorada Gold Project, S-K 1300 Technical Report Summary and Feasibility Study, for Aura Minerals Inc. Effective and report date 31 Dec 2025. §13.5 Table 13-8, printed p.222; Table 12-4, printed pp.173–174.
• Ausenco Engineering Canada et al. (2021) NI 43-101 Technical Report & Feasibility Study on the Las Chispas Project, Sonora, Mexico, for SilverCrest Metals Inc. Effective 4 Jan 2021. §16.5–16.9.
• BBA, InnovExplo et al. (2023) NI 43-101 Technical Report, Feasibility Study for the Cariboo Gold Project, for Osisko Development Corp. January 2023. §16.9–16.11.
• Ivanhoe Mines (2022) Platreef 2022 Feasibility Study (NI 43-101), March 2022. §16.2–16.5.

Cited from verified metadata only (full text not read; flagged on air)

• Vayenas, N. & Wu, X. (2009) "Maintenance and reliability analysis of a fleet of load-haul-dump vehicles in an underground hard rock mine." International Journal of Mining, Reclamation and Environment 23(3):227–238. DOI: 10.1080/17480930902916494. Full text requested.
• Topal, E. & Ramazan, S. (2010) "A new MIP model for mine equipment scheduling by minimizing maintenance cost." European Journal of Operational Research 207(2):1065–1071. DOI: 10.1016/j.ejor.2010.05.037. Full text requested. Described on air only through Fu et al. (2014).
• Topal, E. & Ramazan, S. (2012) "Mining truck scheduling with stochastic maintenance cost." Journal of Coal Science and Engineering (China) 18(3):313–319. DOI: 10.1007/s12404-012-0316-4. Full text requested. Described on air only through Fu et al. (2014).
• Chaowasakoo, P., Seppälä, H. & Koivo, H. (2018) "Age-based maintenance for a fleet of haul trucks." Journal of Quality in Maintenance Engineering 24(4):511–528. DOI: 10.1108/JQME-03-2017-0016. Paywalled.
• Kesimal, A. & Bascetin, A. (2000) "Replacement study of off-highway trucks in an open-pit coal mine in Turkey." Mineral Resources Engineering 9(2):279–286. DOI: 10.1142/S0950609800000226. Full text requested. Crossref gives the issue as 9(2), while the request list had 9(3).
• Siva Prasad, B.N.V., Murthy, V.M.S.R. & Naik, S.R. (2022) "Challenges in drill equipment selection vis-à-vis underground excavations — a methodology." In Lecture Notes in Civil Engineering, Springer, pp.167–182. DOI: 10.1007/978-981-16-9770-8_9. Paywalled. Note the third author is Naik, not Pandey as in the 2016 substitute.
• Sandvik Rock Excavation Handbook: no authorised copy obtained. Not cited for any number.
• UMaT theses (University of Mines and Technology, Ghana): the repository was unavailable on 24 Oct 2026. Not cited for any number.

Cited second-hand (via a paper we did read)

• Trout (1995), Mount Isa MIP schedule, +23% NPV and not implemented; Topal (2003) at Kiruna: both via Little et al. (2008).
• White et al.: the DISPATCH system's 7–20% at 13 mines, via Nehring et al. (2010b).
• Moore (1998): +1% availability gives +1.7–3.5% profit, via Paraszczak (2000). Not used on air.
• Toivanen (1989) and Harjunpää (1988): Finnish LHD cost and overhaul studies, via Matikainen (1991).
• Kuruppu & Golosinski (1998): the MMH/OH benchmark and 80% PM share, via Paraszczak (2000).

Glossary

• Mechanical vs operational availability: Mechanical counts only active repair and PM time as downtime. Operational also counts waiting for parts, fitters and transport. Operational is the planner's number.
• Utilisation / use of availability: The fraction of available time the machine is actually working.
• MTBF / MTTR: Mean time between failures and mean time to repair (or restore; definitions vary).
• MMH/OH: Maintenance man-hours per operating hour, the "maintenance ratio".
• Weibull distribution: A flexible failure-time model whose shape parameter shows whether failure risk rises (above 1), stays constant (1) or falls (below 1) with age.
• Power law process: A model for a repairable system whose failure intensity changes over its life. It is used when failure data show a trend.
• Trend test (Laplace, TTT plot): Checks whether times between failures are getting shorter or longer before you fit a fixed distribution.
• Match factor: The ratio of truck arrival capacity to loader service capacity. 1 is balanced; below 1 is under-trucked.
• MIP: Mixed-integer programming, an optimisation method where some decisions are whole numbers or yes/no.
• Reference-class forecasting: Estimating from the distribution of outcomes in comparable past cases, then adjusting for genuine differences (Flyvbjerg's "outside view").
• Taylor's rule: An empirical relationship between a deposit's tonnage and its sustainable mining rate, fitted on operating mines.

Episode built with the make-a-podcast skill. The script, fact-check and audio were generated by Claude, and every number was verified against the source PDFs. Pronunciations of several proper names (Flyvbjerg, Paraszczak, Matikainen, Booysendal, Sungun, Bascetin, Chaowasakoo, Topal) are best guesses.