Key Takeaways:
Not all excavation tickets carry the same risk. Data-driven prioritization helps utilities identify which ones need closer attention.
Risk scoring combines ticket details, historical damage records, excavator history, and other available data to assess incident likelihood and impact.
Risk scores help teams make better decisions about ticket review, resource assignments, and preventive action.
Picture your ticket queue first thing Monday morning. Hundreds of new 811 tickets, each with a clock ticking. Most are routine. A few could put your most critical infrastructure at risk. The problem is that on paper, they all look pretty similar.
That’s the everyday challenge of excavation ticket management at scale. Tickets usually move through in the order they arrive, sorted by the deadline. That tells you what work is being proposed, but not which jobs need your most experienced people first.
Data-backed risk prioritization fills that gap. It looks at each ticket alongside historical damage records and other data, spots patterns tied to higher damage risk, and helps you focus on the tickets that matter most, faster, before anyone starts digging. You still meet every required response. You just know where to look first.
Risk scoring starts with data. A ticket tells you about the proposed work, but you need more information to judge how likely damage is. Depending on what’s available and the platform you use, a risk assessment can draw on:
Excavation characteristics: the type of work, proposed methods, work dates, and other ticket details
Location information: geographic details that show where the digging will happen and whether that area has a history of damage
Excavator history: records of previous excavation activity, damage incidents, and related patterns
Underground infrastructure: nearby facilities, including asset types, locations, and characteristics, where that data is available
Historical damage records: past incidents that reveal patterns tied to particular locations, activities, or excavators
Any one of these only tells part of the story. A particular excavation method, for instance, isn’t necessarily risky by itself. But when combined with the right location and excavator history, it can signal a much higher chance of damage.
You want to evaluate the relationships among these factors rather than decide based on any single characteristic.
Once the data’s in place, excavation analytics can evaluate it to find patterns associated with previous damage incidents.
Older, rules-based systems check tickets against predefined conditions. You might, for example, flag certain work types or locations for extra review.
More advanced approaches use machine learning models trained on historical excavation and damage data. These models analyze multiple variables at once to find combinations of characteristics associated with a higher likelihood of incidents.
Here’s what that looks like in practice. Your historical data might show that one type of excavation rarely causes damage. But when that same work happens in a particular location, or involves excavators with certain histories, the odds of damage climb. A model can find those relationships across huge datasets, then apply what it’s learned to new tickets as they come in, producing a consistent way to compare the relative risk of different excavation activities.
Identifying these patterns can make a significant difference. In one analysis, Irth found that its predictive risk model identified 52% of damages within the highest-risk 10% of excavation tickets using anonymized data from 811 call centers. With client-specific data, that figure increased to 53%.
A risk score is an estimate, not a prediction that damage will or won’t happen.
Irth’s risk scoring technology, for example, gives each ticket an incident likelihood percentile score, so you can see which excavations rank higher in relative risk according to the model. A higher percentile means the ticket is considered more likely to result in an incident than tickets with lower scores. That isn’t the probability of damage. A 90 doesn’t mean a 90% chance of a strike. It means the ticket ranks high in the model’s assessment.
Good risk scoring also shows its work. When you can see why a ticket got its score, it’s much easier to judge whether the result makes sense and what to do about it.
It’s also worth separating the likelihood of damage from its possible consequences. An excavation may have a relatively low estimated likelihood of an incident but still deserve special attention because of what’s buried nearby or how bad the impact of damage would be. Both considerations are critical in your excavation risk management decisions.
The value of scoring is easiest to see when two tickets look alike. Consider these two hypothetical 811 tickets.
Ticket A is routine residential work. A contractor submits a ticket for a residential utility installation. Your records show few damage incidents tied to similar work in that area, and nothing in the excavator’s history raises a flag. The model gives it a relatively low score.
Ticket B looks similar but carries more risk indicators. Another contractor submits a ticket for the same kind of installation. But this one’s in an area with a history of underground damage, and the records show patterns associated with higher incident risk. The model gives it a higher score.
Both tickets still follow applicable requirements and procedures. But Ticket B might call for a closer look, a conversation with the excavator, or some extra precautions. Without scoring, you’d probably never spot the difference in a long queue.
Now multiply that by a morning’s worth of tickets. In a queue of 500, scoring might surface a few dozen that deserve experienced locators or a site visit, while the rest keep moving through normal workflows within their deadlines. Nothing gets skipped. Your team just knows where to prioritize.
A score on its own doesn’t do much. It matters when it changes what your team does next. A common setup is a risk threshold that triggers extra review, and from there scores can help you:
Flag higher-risk tickets for supervisor review before field work begins
Assign experienced locators or specialized resources to the riskiest jobs
Reach out to the excavator for a call or pre-dig meeting
Spot recurring patterns by location, work type, or excavator that call for outreach and education
These decisions should account for more than the score alone. Regulations, response deadlines, asset characteristics, operational requirements, and professional judgment all still matter. Risk scoring informs those decisions rather than replacing them.
NiSource provides a real-world example. The utility uses Irth Insights to evaluate excavation tickets based on risk and potential impact, helping identify activities that warrant additional attention. Those insights inform targeted communications with excavators and homeowners, along with follow-up from damage prevention specialists when appropriate. Using Irth’s AI-driven impact model, NiSource reported a 290% improvement in identifying critical excavation safety events.
A score is only as useful as the data behind it, the model that produces it, and the way people read it. A few things matter most:
Data quality and completeness: Missing or inaccurate ticket details, incomplete damage histories, and outdated records weaken assessments.
Relevant history: Models need data that reflects the work and conditions it’s judging. What holds in one environment may not hold in another.
Ongoing model evaluation and refinement: Comparing scores with actual damage outcomes shows how well a model separates higher-risk from lower-risk work, and models may need updating as conditions change.
Explainability: Knowing why a ticket scored high helps your team decide whether the result makes sense and what to do.
Human review: Analytics can’t account for every circumstance. Experienced damage prevention staff still need to weigh information the data doesn’t capture.
In the end, a score earns its place when your team can trust it, explain it, and act on it with confidence.
Excavation ticket prioritization lets you size up potential threats using more than what’s on the face of a ticket. By combining ticket details, historical damage records, and other data, excavation analytics can find patterns tied to higher incident likelihood and turn them into scores that show where closer review, preventive action, or specialized attention belongs.
For damage prevention leaders, that means working smarter as volumes grow: putting effort where it counts and making decisions before damage happens, not after.
Irth Insights brings AI-powered risk scoring into excavation ticket management, helping you identify higher-risk tickets, understand what’s driving their scores, and build those insights into your damage prevention workflows.
Utilities can use excavation analytics to evaluate incoming 811 tickets alongside historical damage records, excavator history, location information, and other available data. Machine learning models identify patterns associated with increased damage risk and generate scores that help teams determine which excavations may need closer review or additional precautions.
Utilities can use automated ticket screening and risk scoring to help manage large volumes of excavation notices. Risk scoring identifies tickets with elevated damage potential so teams can focus additional attention and resources where they’re most needed while continuing to meet required response deadlines.
Ticket screening helps determine what action is required for an incoming excavation notice, such as additional review or a field locate. Risk scoring evaluates the relative likelihood of damage to identify tickets that may need more attention. The two processes work together to support more informed excavation ticket management decisions.
Risk scoring can help utilities identify potentially hazardous excavation activity before work begins. By flagging higher-risk tickets, utilities can take preventive measures such as contacting excavators, assigning experienced personnel, or conducting additional reviews. While scoring doesn’t eliminate damage risk, it helps teams make more informed decisions about where to focus their prevention efforts.