Geological Engineering
The Geological Engineering domain of the PEC EPE tests your ability to turn field and laboratory geology into design numbers for foundations, slopes, tunnels, dams, and groundwater control. Questions are practical: given a boring log, core run, or discontinuity set, choose the parameter, classification, or failure mode that governs design.
Core rules and concepts
- Site investigation follows a sequence: desk study and remote sensing, geological mapping, boreholes and trial pits, in-situ tests (SPT, CPT, plate load, packer/Lugeon, pressuremeter), laboratory tests, and monitoring. Depth must pass all significant compressible or weak zones and any karst or cavity.
- Rock mass classification is not a design result by itself. RQD is the percentage of intact core pieces longer than 100 mm in a run; total core recovery is different. Bieniawski's RMR combines UCS, RQD, discontinuity spacing, discontinuity condition, groundwater, and an orientation adjustment. Barton's Q = (RQD/Jn)(Jr/Ja)(Jw/SRF). Hoek's GSI estimates rock-mass strength from blockiness and joint condition.
- Strength for soil uses effective stress: τ = c′ + σ′ tan φ′. For jointed rock, use Hoek–Brown with GSI, disturbance D, and intact UCS; do not insert laboratory intact-rock cohesion directly into a rock-mass model.
- Permeability uses Darcy's law, q = kiA; aquifer tests are interpreted with Theis or Cooper–Jacob; groutability depends on joint aperture and connectivity.
- Slope stability compares resisting and driving forces; kinematic admissibility is checked on stereonets before limit-equilibrium analysis.
- Foundations and hazards require bearing capacity, settlement, liquefaction screening, and fault/karst assessment.
Worked example
A 150 cm core run contains 90 cm of pieces longer than 10 cm. RQD = 90/150 × 100 = 60%. With Jn = 6, Jr = 3, Ja = 2, Jw = 1, and SRF = 1, Q = (60/6)(3/2)(1/1) = 15, i.e. good rock (Q = 10–40). The reasoning matters: RQD uses only competent pieces, and Q is a logarithmic classification, so a single poor input can shift the class.
Traps that cost marks
- Reporting total core recovery as RQD.
- Using peak shear strength for a reactivated landslide without residual strength or pore-pressure reduction.
- Applying RMR without the discontinuity-orientation adjustment.
- Treating Q, RMR, or GSI as final design values rather than inputs to empirical or numerical design.