Earned value management (EVM) is used on traffic signal projects because it does something a simple budget tracker can't: it tells you whether the money you've spent has bought you the work you expected. A project that is 60% through its budget isn't necessarily 60% complete. EVM makes that gap visible, early enough to act on it rather than report it at handover.
Bob Panich Traffic Signals applies EVM discipline across infrastructure delivery programmes, from single-intersection upgrades to multi-site corridor rollouts. The method isn't complicated in principle, but it fails in practice when the three core values are poorly defined or inconsistently measured.
The three numbers you can't skip
EVM rests on three values that must be established before any meaningful analysis is possible.
The first is the Planned Value (PV): the budgeted cost of work scheduled to be complete by a given date. On a traffic signal project, this is built from the programme baseline, which assigns a budget to every work package across the delivery timeline. If programme baselining was done poorly, PV figures will be unreliable from day one, and every EVM metric derived from them will be equally unreliable.
The second is the Actual Cost (AC): what has actually been spent to date. This sounds straightforward, but on traffic signal projects it requires careful invoice timing. Invoices for equipment ordered in month two may not be processed until month four. AC must reflect committed costs, not just paid costs, or the numbers will lag reality by weeks.
The third is the Earned Value (EV): the budgeted cost of work actually completed. This is where most teams go wrong. EV is not a percentage guess applied to the total budget. It is the sum of budget values assigned to discrete, verifiable deliverables that have been finished and confirmed. A signal head mounted and terminated earns its budget. A signal head sitting in a compound earns nothing.
Reading the schedule variance and cost variance
Once PV, AC, and EV are established, two variances tell the story.
Schedule Variance (SV) equals EV minus PV. A negative SV means the project has earned less value than it should have by now: work is behind schedule. A positive SV means the project is ahead. On traffic signal projects, negative SV typically appears first at site access stages, particularly when civil works from another contractor are running late, leaving signal installation teams unable to start. This makes site access planning one of the clearest upstream predictors of SV deterioration.
Cost Variance (CV) equals EV minus AC. A negative CV means the project has spent more than the value of work it has completed. This is the classic cost overrun signal. Positive CV means the project is delivering work for less than budgeted, which is worth understanding: it may reflect genuine efficiency, or it may reflect deferred scope that will show up later.
Indices are more useful than raw variances
Raw variances tell you the size of a gap in dollar terms. Indices tell you the rate at which performance is improving or deteriorating, which is more useful for forecasting.
The Schedule Performance Index (SPI) is EV divided by PV. An SPI of 0.85 means the project is earning 85 cents of value for every dollar of work it was scheduled to complete. An SPI below 1.0 warrants investigation. Below 0.80 on a traffic signal project typically means a structural problem: a delayed approval, a resource shortage, or a supply chain hold-up that isn't going to resolve itself without active intervention.
The Cost Performance Index (CPI) is EV divided by AC. A CPI of 0.90 means every dollar spent has returned 90 cents of earned value. CPI below 1.0 on traffic signal projects often traces to scope additions that weren't formally varied, or to rework caused by installation defects identified late. Good defect management discipline reduces CPI erosion. It's worth reviewing NCR closeout practices alongside CPI trends, because the two are frequently correlated.
Forecasting completion with EVM
EVM doesn't just describe the past. It generates a forecast called the Estimate at Completion (EAC).
The most common EAC formula divides the total Budget at Completion (BAC) by the current CPI. If the BAC is $1.2 million and the CPI is 0.88, the EAC is approximately $1.36 million. That's a 13% cost overrun forecast if current performance continues unchanged. The project manager's job is to decide whether the CPI is likely to improve (because a one-off problem has been resolved) or whether it reflects a structural inefficiency that will persist.
A second formula multiplies the remaining work by both SPI and CPI, which gives a more pessimistic forecast when the project is behind on both dimensions. This composite formula is worth using when the project has sustained underperformance across two consecutive reporting periods. A single bad month rarely justifies a full reforecast. Two months in a row does.
Common mistakes on traffic signal projects specifically
Several EVM pitfalls appear consistently on traffic signal programmes.
The first is treating milestone completion as a proxy for earned value. A milestone is a date, not a deliverable. EV must come from work packages with defined outputs. "Controller hardware installed at site 4" is a work package. "Design complete" is a milestone. Only the former earns value.
The second is using different cost bases for PV and AC. If PV was built on direct labour rates and AC is drawn from invoices that include overheads and preliminaries, the variance figures are meaningless. The cost base must be consistent across all three values.
The third is reporting EVM data without acting on it. EVM exists to trigger decisions: accelerate a work package, raise a variation, challenge a subcontractor's progress claim. Project managers who present SPI and CPI in monthly reports without linking them to specific actions are using EVM as documentation rather than as a management tool.
Setting up EVM on a new project
EVM requires a Work Breakdown Structure (WBS) with assigned budgets before the project starts. On a traffic signal project, a typical WBS separates civil works, equipment supply, installation, testing and commissioning, and documentation. Each element carries a budget percentage and a schedule position.
The budget assigned to each WBS element is the basis for both PV (when it's scheduled) and EV (when it's earned). Without a properly loaded WBS, EVM is guesswork with spreadsheets.
Bob Panich Traffic Signals establishes the WBS and performance measurement baseline at the same time as the project programme. The two documents must be internally consistent. A programme that shows commissioning starting in week 14 must have its commissioning budget loaded to week 14 in the PV curve, or the SV figures from week 14 onward will be distorted.
EVM scales to fit the project. A single-intersection upgrade doesn't need 40 work packages. Five or six well-defined elements, each with a clear completion criterion, are enough to generate meaningful performance data across a short delivery window. The discipline matters more than the complexity.