The FEE syllabus spans engineering mathematics, statics and dynamics, mechanics of materials, fluid mechanics and thermodynamics, circuits, and economics and project management, so a single preparation habit has to work across very different frameworks. This guide trains exactly that habit: before any algebra, name the governing model and its assumptions, then verify the result. You will find a problem-to-model decision table, worked scenarios in three areas, a numerical verification self-check, and a rotating revision plan across all six areas. For administrative matters such as fees, application windows, and format, rely on the Professional Engineers Board (PEB) website rather than third-party summaries.
Turning a worded FEE problem into a named model before computing
Across the six FEE areas, the first decision is classification, not arithmetic. Extract the quantities, units, and unknown, list candidate governing equations, and reject each one whose assumptions the problem statement violates.
A reliable intake routine has four steps: read twice, underline knowns and unknowns with units, sketch the system, then list two or three candidate equations with the assumption each one requires (steady state, rigid body, incompressible flow, linear elements). Only after this do you compute. The routine converts a vague paragraph into a labelled diagram and a single chosen equation, which is what the subsequent algebra actually needs to be correct.
The decision table below doubles as a classification drill. Exercise: take five practice problems from two different areas and, before computing anything, write the model name, the governing equation, and one alternative you rejected along with the assumption it violates. Expected observations: at least one candidate equation per problem fails an assumption check, and stating assumptions in words takes under a minute once practised. Self-check rubric per problem: two points for the correct model, two for a stated assumption, one for a named rejected alternative; aim for four or better before allowing yourself to compute.
| Problem cue | Candidate model | Key assumption | Quick check |
|---|---|---|---|
| Body at rest or constant velocity | Statics equilibrium: sum of forces and moments equal zero | Rigid body; every force identified on the free-body diagram | Sum forces in two directions and moments about a point not on a force line |
| Accelerating body | Newton's second law with kinematics | Consistent inertial frame; mass known and constant | Kinematic relation must match the chosen coordinate direction |
| Fluid flowing steadily through a device | Control-volume mass and energy balances | Steady flow; uniform properties at inlets and outlets | Mass in equals mass out; every energy term in the same units |
| Sinusoidal source, steady state | Phasor circuit analysis | Single frequency; linear circuit elements | Impedances combine like resistances using complex arithmetic |
| Costs and revenues spread over years | Present worth or equivalent annual worth | Stated interest rate; cash-flow timing defined | All alternatives converted to the same time basis before comparing |
Engineering mathematics: picking a numerical method and proving it behaved
Numerical-method questions test selection and verification, not only computation. Match root-finding, integration, or time-stepping schemes to the equation type, then demonstrate convergence by refining the step size or tolerance and checking the residual.
Selection logic is part of the answer. Bracketing methods such as bisection converge predictably but slowly; open methods such as Newton-Raphson converge quickly from a good starting guess but can fail when the derivative vanishes or the guess is poor. For integration, trapezoidal and Simpson rules differ in how error shrinks as the step halves; for differential equations, simple explicit steps need smaller intervals than higher-order schemes to reach the same accuracy.
Verification closes every numerical answer. Substitute a computed root back into the original function and report the residual; halve the integration or time step and confirm the answer stabilises to your stated tolerance; for iterative methods, record the number of iterations and report the residual or the change between successive answers to the tolerance you stated. Exercise: solve one transcendental equation by two different methods. Expected observation: Newton-Raphson needs a handful of iterations from a close guess while bisection predictably halves the bracket each step; writing that comparison down is itself the lesson about method trade-offs.
Statics and dynamics: the moment equilibrium stops, inertia begins
Statics applies when acceleration is zero; dynamics replaces that condition with force equals mass times acceleration and adds a kinematics layer. Classify the motion first, then choose equations, because misclassification produces equations that look right and are not.
Statics discipline means a complete free-body diagram: every contact force, weight, applied load, and couple, drawn before any equation is written. Check for two-force members, keep units consistent in newtons and metres, and fix a sign convention for moments. A diagram missing one support reaction will yield a solved-looking system that fails a total-force check, which is why the check belongs in the written solution, not after it.
Dynamics adds a second decision before the force balance: which kinematic relation describes this motion (constant acceleration, circular, projectile)? Consider a block sliding on a rough plane after a push. A plausible mistake is applying static friction equations while the block still moves; the better decision is to use kinetic friction during motion, then re-examine the body once it stops, comparing the force needed to restart motion against the maximum available static friction. The classification changes twice in one problem, and each switch changes which coefficient and which equilibrium condition apply.
Mechanics of materials: combining axial, bending, and torsional stress at one point
Failure criteria operate on the complete stress state at a point, not on separate component maxima. Compute each stress component with its sign and location, superpose components at the same point, then apply principal stresses or a von Mises equivalent.
Distinguishing the stress types matters because their maximum locations differ. Axial load gives uniform normal stress; bending gives normal stress peaking at the outer fibres with a sign that depends on which side of the neutral axis you evaluate; torsion gives shear peaking at the outer surface; transverse shear peaks at the neutral axis. The worst location for one component is not the worst for another, so the combined check must be made at a single, deliberately chosen point.
Worked scenario: a solid circular shaft carries axial tension plus a torque. A plausible mistake is adding the axial stress magnitude and torsional shear magnitude into one number, or comparing each component separately to its own allowable value. The better decision: compute the normal stress at a surface point, the torsional shear at that same point, combine them into principal stresses or a von Mises equivalent, and compare that single equivalent against the material criterion. This matters because principal stresses depend on how the components orient relative to each other; comparing separate maxima can overstate or understate the combined severity depending on signs and geometry.
Fluid mechanics and thermodynamics: one energy equation, disciplined sign bookkeeping
Both subjects rest on mass and energy balances over a control volume. Define the boundary, list every term with a fixed sign convention, keep units consistent per unit mass or per second, then compute devices such as pumps or turbines.
The modelling choice between equations is the real work. Bernoulli is the mechanical-energy balance along a streamline for steady, incompressible, frictionless flow with no machines; the steady-flow energy equation generalises it with internal energy, heat transfer, and shaft work. Compressibility is a stated assumption, not a default: treating a gas as incompressible is defensible only when pressure and velocity changes are small relative to absolute pressure, and the problem must support that.
Worked scenario: a pump raises water through a pipe with friction losses. A plausible mistake is dropping the friction term or mixing units among metres of head, kilopascals, and kilojoules per kilogram, producing pump-power answers off by large factors. The better decision: write the energy equation between the two free surfaces, carry every term in metres of head, insert friction as an additional head-loss term, convert pressure differences consistently, and multiply by density, gravity, and flow rate for power. This matters because each step is dimensionally checkable, and that check exposes any dropped term immediately.
Electrical circuits: choosing phasors, Thevenin equivalents, and reference directions
Circuit questions reward choosing the right representation: differential equations for transients, phasors for single-frequency steady state, and Thevenin or Norton equivalents when one element's behaviour is of interest. Fix reference directions before writing any equation.
The phasor domain converts integro-differential equations into complex algebra: a resistor keeps its resistance, an inductor becomes j-omega-L, a capacitor becomes one over j-omega-C. This shortcut is valid only at a single frequency in sinusoidal steady state, so checking that condition is part of choosing it. Thevenin analysis proceeds by finding the open-circuit voltage and the equivalent impedance seen from the terminals, with sources deactivated correctly, so any load can then be attached without re-solving the whole network.
Reference directions and the passive sign convention are bookkeeping, but errors here corrupt power calculations. Choose current directions and voltage polarities at the outset; a negative answer then carries meaning about direction instead of confusion. Use power conservation as a built-in verification: total power delivered must equal total power absorbed. Exercise: reduce a two-source resistive network to its Thevenin equivalent, attach three different loads, and confirm load voltages against a direct nodal solve. Expected observation: both routes agree, and the equivalent approach scales better whenever the load changes.
Economics, project management, and a rotating revision plan across all six areas
Engineering economics compares cash flows only after moving them to a common time basis, and project management questions test reading activity networks and float. Combine both with a revision rotation that revisits every FEE area weekly.
Worked scenario: choosing between two machines, one cheaper to buy with higher yearly operating costs. A plausible mistake is ranking by purchase price alone. The better decision: compute a present worth or an equivalent annual cost for each alternative at the stated rate, including salvage values, then compare like with like. This matters because cash-flow timing changes rankings; the cheaper machine can genuinely lose once every cost sits on one common basis. The same structure covers replacement analyses and comparisons of alternatives with unequal lives.
On the project management side, distinguish the critical path, which is the longest path through the network and sets the minimum project duration, from float, which is the scheduling slack on non-critical activities. A forward and backward pass yields earliest and latest start and finish times for each activity. Exercise: build a small activity network and compute these by hand. Expected observation: activities on the longest path show zero float, and the total float of a non-critical activity equals its latest start minus its earliest start, which confirms your pass arithmetic.
- Rotate through all six areas on a weekly cycle so no area sits untouched for more than a week, weighting time toward areas where your self-check rubric scores are lowest.
- Adaptable preparation sequence: first consolidate fundamentals (mathematics and numerical methods, statics), then practise combined-topic problems such as stress superposition and control-volume balances, then attempt timed mixed sets drawn across all areas.
- Readiness check one: for a fresh problem in each area, you can write the model name and one key assumption before computing anything.
- Readiness check two: your numerical answers include a convergence or dimensional verification, not just a final number.
- Readiness check three: your written solutions name rejected alternative models and the assumption that disqualified them, evidence that the classification step is active rather than skipped.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.