Near Misses and System Failures: Why Modern Safety Depends on Updating Old Infrastructure

But a near miss is not simply an avoided accident. It reflects a situation where normal safeguards were stretched or bypassed, even if only briefly. When similar incidents repeat, they begin to form a pattern that reveals how a system behaves when pressure increases.

In a recent post referencing his article Too Many Near Misses — The Urgent Need to Modernize Pilot Training and a recent post, entrepreneur and jet-rated pilot Sky Dayton argued that mounting near misses should not be dismissed as isolated anomalies but recognized as signals that core systems need updating. The point extends well beyond aviation. Repeated close calls often indicate structural strain long before institutions acknowledge it.

Many organizations treat near misses as isolated problems rather than systemic indicators. The reasoning feels logical. If nothing failed, then the controls must have worked. Yet that logic overlooks how often stability depends on human intervention rather than reliable design. Skilled operators fill gaps, adapt processes, or make judgment calls that prevent escalation.

Over time, reliance on human adaptation can mask structural weaknesses. Systems appear stable because people compensate for them. The absence of visible failure creates a false sense of security, allowing risk to accumulate in the background.

Systems Built for a Different World

Large infrastructure systems rarely begin with flawed intentions. They are designed around the realities of their time, including expected demand, available technology, and known risks. Problems emerge when those assumptions remain fixed while operating conditions change.

Equipment that once operated independently may now exchange data constantly with other platforms. Maintenance cycles that worked under predictable usage patterns can struggle when workloads fluctuate or increase unpredictably. The result is not immediate failure but gradual strain that shows up in subtle ways, such as increased troubleshooting or growing reliance on manual checks.

The condition of national infrastructure reflects that long-term drift. In its most recent national assessment, the American Society of Civil Engineers assigned U.S. infrastructure an overall grade of C, a finding detailed in the ASCE Infrastructure Report Card release. A midrange grade does not suggest imminent collapse, but it does indicate systems that are aging, strained, and increasingly dependent on sustained maintenance rather than modern redesign.

Legacy software environments illustrate a parallel shift. Many organizations continue to rely on systems that have been extended repeatedly through updates rather than replaced. Each new layer solves a specific need, yet the overall architecture becomes harder to understand. Engineers may know how individual components work but struggle to see how everything interacts under stress.

Institutional structures often slow adaptation. Procurement timelines favor incremental upgrades instead of fundamental redesign. Policies written for earlier technological environments may not reflect modern risks. None of these factors cause immediate harm, but together they create a widening gap between how systems were intended to operate and how they actually function.

The Warning Signs We Keep Ignoring

Near misses produce valuable operational data, yet that data does not always lead to meaningful change. Reports may exist in internal systems, but they often become routine paperwork rather than catalysts for improvement. When incidents are categorized as minor deviations, they lose urgency.

Research examining the relationship between workplace accidents and near misses has found that near misses occur far more frequently than formal accident reports, with one large-scale analysis identifying roughly 2.54 near misses for every recorded occupational accident, a ratio detailed in a peer-reviewed study published through the National Library of Medicine’s PubMed Central archive. That imbalance reinforces how much potential failure activity takes place before visible harm occurs.

Psychology plays a role in how those warnings are interpreted. When teams experience repeated close calls without negative outcomes, expectations shift gradually. Practices that once felt risky can begin to seem routine because they appear manageable. The threshold for concern rises quietly, even as exposure increases.

Organizational incentives also influence whether near misses receive attention. Escalating concerns can trigger operational slowdowns or additional oversight. Individuals may hesitate to frame an incident as systemic if doing so invites scrutiny. The result is that warning signals accumulate without prompting structural correction.

Technology has increased the volume of available monitoring data, yet more information does not automatically produce action. Dashboards may highlight irregularities continuously, creating background noise that makes it harder to distinguish genuine risk from routine fluctuation. Without deliberate analysis, near misses remain isolated entries rather than early indicators of deeper strain.

When Redundant Isn’t Reliable

Adding backups is a common response to perceived risk. Redundancy feels reassuring because it suggests there is always another layer of protection. However, redundancy without thoughtful redesign can introduce complexity that makes systems harder to manage.

Engineering research has shown that even highly redundant systems remain vulnerable to common-cause failures that can undermine multiple safeguards at once. Analysis summarized in NASA’s technical reliability research archive notes that shared failure mechanisms can limit the effectiveness of redundancy, highlighting how layered protections do not always translate into independent safety margins.

Over time, additional safeguards can create intricate webs of dependencies. Operators must understand not only how primary systems work but also how backup mechanisms interact with them. During normal operations, this complexity may remain invisible. Under pressure, it can slow response times or create confusion about which controls take priority.

Layered technology environments highlight this challenge. Systems that began as simple architectures evolve through incremental additions. Interfaces, patches, and compatibility layers allow old and new components to coexist, but they also make troubleshooting more difficult. A problem that appears isolated may actually stem from interactions between multiple generations of technology.

Redundancy also changes human behavior. Teams may assume that backup systems reduce the need for proactive maintenance or design improvements. Over time, this mindset shifts attention away from underlying issues. Redundancy can prevent immediate failure while allowing deeper problems to remain unresolved.

The Spreadsheet Problem: Why Maintenance Rarely Wins

Maintenance competes for attention in environments where progress is often measured by visible outcomes. New projects generate momentum and recognition. Upgrading existing infrastructure, even when necessary, lacks the same visibility.

Budgeting practices reinforce this imbalance. Maintenance appears as a recurring expense rather than a strategic investment. Decision makers may understand the long-term value of upgrades, yet short-term financial pressures encourage postponement. Each delay increases dependence on workarounds that keep systems operating.

Operational teams often rely on detailed tracking tools to prioritize tasks. These tools provide structure, but they can also narrow focus to immediate concerns. Larger structural issues may remain unresolved because they do not fit neatly into existing planning cycles.

The result is a cycle where systems receive incremental fixes rather than comprehensive attention. Stability becomes a product of constant adjustment rather than intentional design.

Patch Culture and the Illusion of Safety

Temporary fixes rarely remain temporary. When teams develop quick solutions to address urgent challenges, those solutions often become standard practice. Over time, the system’s original design becomes difficult to distinguish from accumulated adjustments.

Software environments make this dynamic especially clear. Patches address vulnerabilities or compatibility issues quickly, which allows operations to continue. Yet each additional layer increases complexity and reduces clarity about how the system behaves as a whole. Operational processes evolve in similar ways. Procedures designed for unusual circumstances may remain in place long after the original context disappears. New staff inherit these practices without understanding why they exist, which makes them harder to question or remove.

Success reinforces the illusion that everything is under control. If systems continue functioning, the growing complexity behind them may go unnoticed. The margin for error narrows slowly, making risk harder to recognize until a larger failure occurs.

Modern Failures Don’t Stay Local

Interconnected systems allow organizations to operate efficiently, but they also increase the speed at which problems spread. When infrastructure components rely on shared data or centralized controls, disruptions in one area can affect multiple operations simultaneously.

Digital integration accelerates feedback loops. Automated processes respond quickly to changing conditions, which can amplify small errors. Operators may encounter cascading effects before they have time to diagnose the initial issue.

Physical infrastructure reflects similar patterns. Systems that once operated independently now share resources or communication channels. This interconnectedness improves coordination but reduces isolation between failures.

Planning for safety requires understanding these relationships. Addressing risk within a single component may not prevent wider disruption if dependencies remain unexamined. Effective oversight must consider how systems interact rather than treating them as separate entities.

What Real Modernization Looks Like

Modernization involves more than replacing aging components. Effective upgrades require examining how systems respond to uncertainty. Instead of assuming predictable operating conditions, designers must consider how infrastructure behaves when assumptions break down.

Stress testing provides insight into these dynamics. By examining performance under unusual scenarios, teams can identify weaknesses that routine operations may hide. These exercises often reveal unexpected dependencies or procedural gaps that would otherwise remain unnoticed.

Modular design offers flexibility. When systems consist of smaller independent components, updates can occur without disrupting the entire structure. Modular approaches also simplify troubleshooting by isolating problems more effectively. And cultural change is just as important as technical upgrades. Organizations that treat maintenance and modernization as continuous responsibilities are better prepared to adapt. Safety improves when teams view updates as normal practice rather than exceptional events.

Learning From Near Misses Instead of Waiting for Failure

Near misses provide an opportunity to strengthen systems before serious consequences occur. Organizations that analyze these events consistently develop a clearer understanding of how their operations behave under stress.

Structured reporting encourages openness. When employees feel comfortable sharing concerns, patterns emerge more quickly. Transparency reduces the likelihood that similar issues will remain hidden across different departments.

Leadership plays a central role in sustaining this approach. Decision makers must prioritize long-term stability even when immediate performance pressures compete for attention. Without visible support, near-miss reporting may decline as teams focus on short-term outcomes.

Improvements based on near-miss analysis often appear incremental, but their cumulative impact can be substantial. Addressing small weaknesses early prevents them from aligning into larger failures.

Safety Is Not a Static Achievement

Infrastructure rarely collapses without warning. More often, systems drift gradually toward instability as conditions change and adjustments accumulate. Recognizing this drift requires paying attention to patterns that develop over time.

Safety depends on continuous attention rather than fixed milestones. Systems that appear stable today may rely on assumptions that no longer hold true tomorrow. Treating infrastructure as finished encourages complacency. Updating infrastructure involves technical decisions, economic trade-offs, and organizational priorities. Progress occurs when leaders accept that maintaining safety requires ongoing effort rather than occasional intervention.

Near misses reveal how systems behave when they approach their limits. Ignoring those signals allows risk to build quietly. Listening to them creates an opportunity to strengthen infrastructure before the next warning becomes something harder to contain.