CSS-Transit common mistakes
CSS-Transit · Understand the traps and how to avoid them.
When a scheduling or headway question gives a peak-period demand figure, the candidate divides the hourly demand by vehicle capacity once and reports that quotient as the required peak headway, skipping the load-factor standard and the round-trip cycle-time check before committing to an answer.
Headway arithmetic feels like a single-step word problem, so candidates stop at the first plausible number. They forget that required vehicles come from cycle time divided by headway, and that a policy load factor (not crush capacity) sets the denominator. The result is an answer that is arithmetically clean but operationally infeasible.
What to do instead
Adopt a fixed three-step template for every headway item: (1) convert demand to design demand using the policy load factor, (2) compute required headway = 60 x design capacity per vehicle / design demand, (3) compute fleet = ceiling(cycle time / headway). Write all three lines even when only one is asked, then select the option that matches step 3.
On disruption and recovery scenarios, the candidate recommends the first mitigation that restores service (short-turning, adding buses, or suspending a segment) without first stating how riders are protected, whether the mitigation preserves transfer connections, and what triggers returning to normal service.
Exam stems are written to reward structured incident command thinking, but under time pressure candidates pattern-match to the most familiar tactic. They treat service restoration as the objective rather than one constraint among safety, rider impact, and labor/vehicle availability.
What to do instead
Before answering any disruption item, write four labels in the margin: safety, rider impact, resource feasibility, recovery trigger. Evaluate each candidate option against all four and eliminate any that ignores rider protection or lacks a defined exit condition. The correct option almost always names both the interim measure and the condition for ending it.
The candidate answers safety-risk questions by choosing the control that eliminates the hazard outright, rather than ranking controls in the hierarchy (elimination, substitution, engineering, administrative, PPE) and selecting the highest feasible level for the described exposure.
Candidates conflate 'safest sounding' with 'correct according to the hierarchy of controls'. They also forget that training and signage are administrative controls and therefore rank below physical separation, barriers, or design changes.
What to do instead
Memorize the five-level hierarchy as an ordered list and, for each option, tag its level explicitly. If two options are both feasible, the higher level wins. When the stem says the hazard cannot be removed, the answer is the highest remaining level, typically an engineering control such as platform edge treatment or signal enforcement.
When a scenario involves a security threat or emergency, the candidate jumps to the tactical response (evacuation, lockdown, police notification) and never addresses the preparedness cycle elements the question is testing: risk assessment, plan currency, training and drills, and after-action review.
Dramatic scenarios pull attention to the moment of response. Candidates also assume 'security' means law enforcement action, missing that transit security management is largely preventive and documented.
What to do instead
Read the stem for the phase being tested (prevent, protect, respond, recover) and answer only within that phase. If the stem describes a gap discovered during an event, the correct answer is usually a preparedness or corrective-action element, not a tactical one. Keep a one-line definition of each phase and check your choice against it.
The candidate treats standards and regulatory items as general knowledge and picks the answer that sounds most rigorous, without distinguishing mandatory requirements (regulations, adopted codes) from voluntary consensus standards, guidance, or recommended practice.
Many transit documents look alike in tone. Candidates do not habitually ask 'is this enforceable, and by whom?' so they select a voluntary standard where the stem asked what an agency must do, or vice versa.
What to do instead
For every compliance item, classify the cited instrument on two axes: mandatory versus voluntary, and internal policy versus external requirement. Answer according to the classification the stem implies. If the stem uses 'must', 'required', or 'shall', eliminate voluntary guidance options immediately.
On procurement questions, the candidate selects the lowest-price or technically strongest proposal without checking whether the solicitation method, evaluation criteria, and protest/appeal steps were followed in the correct sequence.
Candidates evaluate the outcome (who should win) instead of the process (was the process lawful and consistent with the stated criteria). Public procurement rewards procedural regularity, and exam items test exactly that.
What to do instead
For each procurement item, reconstruct the sequence: need definition, method selection, solicitation, evaluation against published criteria, award, debrief, protest window. Identify which step the stem is really about and answer there. Never let price or technical merit override a process defect described in the stem.
The candidate answers technology questions by choosing the most capable or newest system (for example, a full communications-based train control migration) rather than matching the technology to the stated operational need, existing infrastructure, and lifecycle cost.
Vendors and headlines make advanced technology salient. Candidates also underweight integration and migration risk, so they pick an option that cannot be deployed on the described asset base or within the stated budget horizon.
What to do instead
For each technology option, write the operational need it addresses and one integration constraint it imposes. Eliminate options that exceed the need or conflict with the described legacy systems. Prefer the option that delivers the required function with the smallest change to proven, supported infrastructure.
When a data or systems scenario involves automatic vehicle location, fare collection, or predictive maintenance, the candidate answers as if data were automatically trustworthy and skips data quality, validation, privacy, and cybersecurity considerations named or implied in the stem.
Candidates focus on the analytics payoff and treat the data pipeline as a given. Transit technology items frequently embed a data-integrity or security flaw that changes the correct answer.
What to do instead
Before selecting, scan the stem for words like 'incomplete', 'delayed', 'shared', 'third-party', or 'personal'. If present, the correct answer usually includes a governance step (validation, access control, anonymization, retention) rather than a pure analytics step. Add a mental checklist: source, quality, security, privacy, use.
The candidate answers workforce and leadership items with generic management platitudes ('communicate better', 'provide training') instead of naming the specific supervisory or labor-relations mechanism the situation calls for, such as a documented performance improvement process, a joint labor-management committee, or a structured mentoring and certification path.
Leadership questions feel subjective, so candidates retreat to safe-sounding generalities. They also overlook that transit workforce items are governed by collective bargaining agreements, certification requirements, and defined progression paths.
What to do instead
Convert each option into a concrete action with an owner, a document, and a timeline. Eliminate any option that cannot be implemented or measured. When a labor agreement is mentioned, prefer options that follow the contractual step (grievance procedure, joint committee) over unilateral management action.
The candidate manages time poorly across the exam: spending excessive minutes on calculation-heavy operations items early, then rushing later scenario questions and changing correct first answers without a stated reason.
Operations and technology items invite verification loops, and candidates lack a pacing rule. Anxiety near the end produces answer-changing driven by doubt rather than new evidence.
What to do instead
Set a per-question time budget before starting (total minutes divided by item count), flag and skip any item exceeding it by 50 percent, and reserve the final block for flagged items only. Change an answer only when you can name the specific fact or calculation that was wrong; otherwise keep the original.