Study the CEM-C by rehearsing scenario reading, not formula recall. For each practice case: name the measure being proposed, identify which loads (baseload or weather-sensitive) it actually affects, then rank options on normalized cost per unit saved rather than headline payback. Practice with metric units and Canadian rate structures, and always state what you would verify before and after implementation. If you can explain why a tempting answer is wrong in one sentence, you are ready for that question type.
Scenario questions bundle several problems into one fact pattern
CEM-C style cases mix data interpretation, measure selection, cost ranking, and documentation in a single stem. Solve them by separating concerns before computing anything.
A typical stem gives you billing data, a proposed retrofit, a budget constraint, and an environmental or code consideration all at once. Train a fixed reading sequence: first, what is the question actually asking you to recommend or decide; second, what numbers are given and which are irrelevant distractors; third, what assumption the question expects you to state. Write this sequence on your scratch paper at the start of practice so it becomes automatic.
The most common self-inflicted error is answering the question you expected instead of the one asked. A stem may present a lighting retrofit with full data while the actual ask is whether the ventilation interaction changes the recommendation. After every practice case, reread the question line and check that your conclusion addresses its verb: recommend, rank, verify, or explain. If your answer does not match the verb, redo the case.
- Step 1: Restate the decision in one sentence before touching numbers.
- Step 2: List given data and mark each item used or unused.
- Step 3: Check your conclusion against the question's action verb.
Baseload versus weather-sensitive load drives which measures matter
Classifying a facility's load as baseload or weather-sensitive determines which efficiency measures will produce savings and which will disappoint.
Baseload is consumption that runs regardless of outdoor conditions: lighting in continuously occupied areas, process equipment, plug loads, and domestic hot water recirculation. Weather-sensitive load tracks heating degree days or cooling degree days: space heating, much of ventilation air conditioning, and freeze protection. A measure aimed at the wrong category yields real unit savings but a negligible share of total cost. Before evaluating any retrofit, sketch the load breakdown and confirm the measure targets the dominant category.
This distinction matters for interpretation questions too. If consumption rose in a mild winter, the cause is more likely a change in baseload equipment, occupancy, or meter error than weather, because weather-sensitive demand should have fallen. In your answer, state the classification explicitly: 'the increase is inconsistent with weather-sensitive load, so investigate baseload equipment and billing.' That sentence converts an observation into a defensible management conclusion.
Normalize consumption before comparing buildings or years
Raw consumption comparisons mislead whenever weather, occupancy, or production differ. Normalization using degree days or production units isolates real efficiency change.
Two buildings with identical floor area can have very different normalized energy use intensity because one is a warehouse and one is a laboratory. Within one facility, comparing last January with this January is unreliable if heating degree days differed. The working habit is to express energy use per unit of driving variable: kWh per square metre per year for buildings, or energy per unit of production for industrial sites. Then a change in the normalized figure is evidence of efficiency change, not just demand change.
Practice reconstruction: given monthly gas use and monthly heating degree days, check whether the ratio moves consistently; where it does not, that month has an anomaly worth explaining. In answers, state which normalization you used and its limits. A candidate who says 'consumption fell 8% year over year' and a candidate who says 'normalized intensity fell 8% after degree-day adjustment' are giving different quality of evidence. The second is the manager's answer.
- Normalize energy per floor area, per operating hour, or per production unit.
- Adjust for heating and cooling degree days before comparing heating or cooling seasons.
- Flag months that break the normalized pattern and seek an operational cause.
Worked scenario one: choosing between two retrofit options
A warehouse case shows why ranking options on cost per unit saved, not headline payback, changes the recommendation.
Fact pattern: a distribution warehouse spends most of its electricity on high-bay lighting running 5,000 hours per year, and its natural gas mostly on unit heaters in a heating-dominated climate. Option A is a lighting retrofit saving 120,000 kWh per year at a cost of 90,000 dollars. Option B is a heating controls retrofit saving 25,000 cubic metres of gas per year at 60,000 dollars. Option A has the more attractive simple payback, and the tempting mistake is to recommend it and move on.
The better decision checks two things. First, which load dominates the bill: in this pattern, heating is the larger cost, so Option B addresses more of total spend. Second, the conversion basis: express both savings in a common unit (for instance, convert gas savings to its electricity-equivalent cost) and compare dollars saved per dollar invested over a stated horizon. The recommendation can still favour Option A, but now it is justified by the normalized comparison and by the interaction note: fewer hours of high-wattage lighting slightly increases heating load, which the lighting-only payback ignores. State both effects and the decision is defensible either way; state only payback and it is not.
Worked scenario two: reading a consumption increase without blaming weather
A hospital case demonstrates how to interpret an unexplained consumption rise by separating baseload, weather, and operational explanations.
Fact pattern: a hospital's monthly electricity use has been flat for two years. This February, consumption jumps 15% over last February, and the facility manager proposes attributing it to a colder month. The tempting mistake is to accept the weather explanation because the month was, in fact, colder. The better analysis checks the classification: hospital electricity is dominated by baseload (ventilation fans running continuously, medical equipment, lighting), which barely responds to outdoor temperature, while the heating side is gas. A 15% electricity rise is inconsistent with a weather-sensitive explanation for an electricity meter.
The defensible next step is a short verification list: confirm the meter reading and billing period alignment, check for added equipment or extended operating hours, review whether any temporary systems (construction heaters, supplemental cooling) were energized, and only then consider weather as a contributing factor for the small weather-sensitive fraction. Write the conclusion as a ranked hypothesis with the evidence that would confirm each. This structure, observation, classification, ranked causes, verification plan, is exactly the pattern to reproduce in exam-style answers and in real audit reports.
Simple payback, lifecycle cost, and rate structure give different winners
The same two measures can rank differently under simple payback, lifecycle cost, and rate-aware analysis. Knowing which lens the question uses prevents wrong recommendations.
Simple payback divides first cost by annual savings and ignores everything after the payback year. Lifecycle (or total-cost) analysis sums costs and savings over the measure's service life, often discounted, and usually favours longer-lived, higher-capital measures. Rate structure matters because savings in peak-demand periods are worth more than savings at off-peak hours under time-of-use billing; a measure saving modest energy at peak can beat a larger off-peak saving. When a scenario specifies cheap off-peak electricity or a demand charge, that detail is a signal to apply rate-aware reasoning.
Before computing, ask which lens the question wants: if it gives maintenance costs, service life, or discount details, use lifecycle reasoning; if it gives rate periods or demand charges, use rate-aware reasoning; if it gives neither, simple payback with an explicit caveat is acceptable. The comparison table below summarizes the lenses. Practise the same fact pattern under all three and watch the ranking flip; that experience is what makes the recognition automatic during the exam.
| Lens | Inputs used | Typical result | Use it when the stem gives... |
|---|---|---|---|
| Simple payback | First cost, annual savings | Fast, low-capital measures win | Only cost and annual savings |
| Lifecycle cost | First cost, savings, service life, maintenance, discount details | Durable, efficient measures win | Service life or operating cost details |
| Rate-aware | Savings by time period, demand charges, rate structure | Peak-shifting measures win | Time-of-use rates or demand charges |
Measurement and verification, documentation, and Canadian context
Every recommendation is stronger with a stated baseline, a defined measurement boundary, and verification terms. Frame practice cases around Canadian units, codes, and climate.
A measurement and verification (M&V) plan says what will be measured, against what baseline, over what period, and who accepts the result. In scenario answers, add one verification sentence per recommendation: 'confirm lighting hours with occupancy or meter data for 60 days before and after.' This converts a claim of savings into a checkable statement, and it is also how documentation questions are scored in spirit: an energy manager's conclusions should be traceable to data, assumptions, and stated uncertainty. When ethics or safety considerations appear (for example, a measure that reduces ventilation below code minimums to save energy), the correct answer flags the constraint rather than the saving.
Canadian practice specifics to rehearse: metric units throughout (kilowatt-hours, litres, cubic metres of gas, square metres), heating-dominated climates in most regions so envelope and heating measures carry weight, electrically heated buildings where demand management matters more, and cold-climate interactions such as heat recovery on ventilation exhaust. When you read a practice case, mentally locate it: a Vancouver office and a Saskatchewan arena will rank measures differently. For administrative details of the credential itself, consult the AEE directly rather than secondary sources.
- State the baseline, boundary, and verification period for every proposed measure.
- Flag any measure that trades energy savings against code, comfort, or safety compliance.
- Rehearse with metric units and heating-dominated and electrically-heated variants.
A self-check rubric and an adaptable preparation sequence
Score every practice case against a five-item rubric, then run a staged sequence from concepts to timed mixed cases until your rubric scores stabilize.
Rubric (score each 0-2; a 7+ on 10 after two weeks of practice is a reasonable learning milestone, not a pass prediction): (1) decision restated in one sentence; (2) load classified as baseload or weather-sensitive; (3) data used with normalization stated; (4) recommendation justified by the correct cost lens; (5) verification or documentation note included. Keep a log of cases and rubric scores; the item that stays low is your weakest concept, not a reason to grind more questions. Pair each weak item with one worked example rewrite: take your answer and rewrite it as the model answer, noting exactly what sentence was missing.
Sequence: weeks one to two, learn the concept pairs (baseload vs. weather-sensitive, payback vs. lifecycle, measurement vs. verification) and do ten single-concept cases, scoring with the rubric. Weeks three to four, do mixed cases with interaction notes and Canadian context, and add the timing habit: one case in eight minutes. Final stretch, full mixed sets under time pressure, then review by rewriting only your two worst answers. On exam day, the reading sequence and rubric run themselves, and each question gets a stated assumption, a normalized comparison, and a verification sentence.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.