Alyeska Pipeline and the operating record behind Alaska's critical infrastructure

Summary

• Alyeska Pipeline Service Company's relevant technology surface is the operating record behind the Trans Alaska Pipeline System: SCADA telemetry, communications systems, pump-station automation, smart-pig integrity data, flow-assurance monitoring, terminal inventories, emergency-response plans and regulatory evidence.
• The public record supports a serious, long-running critical-infrastructure operator, but it does not let an outside reader test private SCADA resilience, cybersecurity controls, alarm quality, data lineage, backup and recovery, internal access policy or incident-ticket performance.
• The strongest evidence comes from Alyeska's public TAPS pages, the 2024 TAPS Factbook, historical throughput data, the 2025 sustainability report, emergency and SERVS disclosures, Valdez Marine Terminal information and independent reporting on right-of-way and climate-review pressure.
• The main risk is not that Alyeska lacks a pipeline. The risk is that low-flow physics, remote field work, aging assets, permafrost and flood exposure, terminal modernization, tanker response, regulatory scrutiny and cyber/OT boundaries all depend on records that must remain fresh, governed, queryable and recoverable.

The technology surface is the operating record

Alyeska Pipeline Service Company is easy to misread if the word "pipeline" is treated as a generic asset label. The company was formed in 1970 to design and build the Trans Alaska Pipeline System, then operate and maintain it after startup. That history matters, but it is not enough for a technology assessment. The current question is not whether a pipe exists from the North Slope to Valdez. It is whether Alyeska's operating system can keep the physical line, the terminal, pump stations, control data, maintenance records, flow-assurance work and emergency plans in agreement while the system is used every day.

The public boundary is clear. Alyeska's own company page describes the organization as the operator and maintainer of TAPS, a system that began moving oil on June 20, 1977 and passed the 19 billion barrel mark in 2025. The TAPS facts page gives the physical envelope: an 800-mile, 48-inch line crossing mountain ranges, major rivers and hundreds of smaller waterways, with valves, vertical support members, right-of-way widths and temperature extremes that turn geography into an operating constraint. Those figures are not just background facts. They explain why Alyeska's real information problem is distributed, remote and unforgiving.

The same public material shows why the company cannot be judged like a cloud software vendor. Alyeska does not publish a commercial data platform for ordinary customers to test. It does not expose a public control API. It does not invite outside users to run workload benchmarks against its SCADA systems, emergency notification workflows or integrity databases. The public can inspect disclosures, regulator-facing descriptions, monthly throughput numbers, sustainability reporting and response-capability claims. It cannot directly test the control room, the historian, the cyber boundary, the failover plan or the field work-order stack.

That limit is not a reason to abandon the analysis. It is the analysis. Critical infrastructure often reveals itself through the evidence it can safely publish and through the gaps it must keep private. Alyeska's public record is rich enough to describe the operating surface: an Operations Control Center, communications links, telemetry, remote pump-station control, low-flow mitigation, smart-pig data, inspection programs, Valdez Marine Terminal inventory and vapor systems, emergency response resources, public drills and regulator correspondence. It is not rich enough to prove private system performance.

A fair assessment has to hold both points at once.

In that sense, Alyeska is a data-infrastructure case because the system's output is not merely transported oil. The output is operational confidence. Operators need to know pressure, flow, temperature, tank levels and valve positions. Engineers need to know whether corrosion, dents, deformation, ground movement or support degradation require intervention. Maintenance teams need to know which assets are in service, on standby, demolished, warming crude or serving as response bases. Emergency teams need to know which equipment, responders, plans and contact points are ready. Regulators need evidence that commitments are being met.

The public needs enough transparency to believe that the line is still being operated prudently.

That is why the assigned technical question - whether the system keeps data fresh, governed, queryable and recoverable under repeated use - is not abstract. A stale pump-station record, a delayed alarm classification, an outdated response cache, a weak inspection lineage or a confused OT/IT boundary can become an operational problem even when the steel pipe is intact. Alyeska's public evidence should therefore be judged by how it reveals control, continuity and uncertainty, not by whether it sounds like a modern software pitch.

TAPS turns geography into a data problem

The Trans Alaska Pipeline System is a physical system with unusual geography. Alyeska's public facts describe an 800-mile route from Prudhoe Bay to Valdez, a 48-inch pipe, three mountain ranges, major river crossings and severe air-temperature extremes. The Valdez Marine Terminal adds a southern endpoint with 1,000 acres on Port Valdez, two loading berths, 13 crude storage tanks in service and 6.6 million barrels of working inventory capacity. This is not a single plant with one data room. It is a distributed operating environment where field context and central records have to converge.

That geography creates the first data-governance requirement: location has to be precise. A pressure variance, corrosion indication, support movement, river-bank concern, spill report, valve action or maintenance job is not useful unless it is tied to the right milepost, asset, pump station, response area and plan. The line crosses permafrost, rivers, mountains, highways and marine interfaces. A record cannot be merely a spreadsheet row called "pipeline issue." It has to be locatable, comparable with older inspections, visible to the right teams and recoverable after a shift handoff or incident review.

The 2024 TAPS Factbook gives one of the clearest signs that Alyeska's operating surface is a communications and control network as much as a pipe. Its glossary says the communications system comprises microwave and fiber. It says control data systems provide supervisory control and telemetry, seismic monitoring, maintenance monitoring and control of pipeline operations, while enterprise data services support voice, data, video and internet for business systems. That split is important. It suggests a distinction between operational control and ordinary enterprise connectivity, a basic requirement for OT/IT boundary management.

The same Factbook describes the Operations Control Center as continually monitoring pump stations and valves using SCADA systems with remote sensors. Pressures, flow rates, temperatures, tank levels and valve positions are recorded and analyzed for abnormal operations or indications of a leak. The controller can change pump speed or relief valve settings and can issue idle or stop commands. That is a clear public description of an operating loop: sense, record, analyze, decide and act. It is also an implicit data-quality claim, because the loop only works if readings are current, labeled correctly and trusted by qualified operators.

The technology question becomes sharper when pump-station status is considered. Alyeska's pump-stations page says the system was originally designed for 12 pump stations, 11 were built, and only four are needed today because of lower throughput and Electrification and Automation upgrades. Pump Stations 1, 3, 4 and 9 currently pump oil through TAPS. Pump Station 5 is a relief station. Pump Station 7 is part of the cold-restart contingency picture. Several former stations have been ramped down, repurposed, cleaned for demolition or demolished.

That is asset lifecycle management in public view. A station can be active, a response base, a relief station, warm standby, demolished or awaiting demolition. Those statuses are not decorative. They affect control-room commands, maintenance schedules, spare parts, field staffing, safety planning, response staging and restart assumptions.

The Factbook adds that Pump Stations 1, 3, 4 and 9 have been upgraded to electrical power and are automated, controlled remotely by the Operations Control Center, and that upgrades included new onsite control, data gathering and data transfer systems plus pressure-protection and process-safety command upgrades. In other words, modernization created a stronger data dependency, not a weaker one.

The public evidence does not show the architecture behind those control and data-transfer systems. It does not identify vendors, network segmentation, historian retention, backup cadence, alarm thresholds, failover design or access-control policy. It does, however, show that Alyeska's operating model depends on a distributed telemetry fabric.

If the public wants to know whether the company is relevant to network-resource evidence, this is the answer: not because Alyeska is an internet carrier, but because an operational communications network, control data systems and field telemetry are essential to the safe movement of a strategic commodity.

Low flow is a continuous operating test

Alyeska's historic throughput page is one of the most useful public data sets because it turns a long infrastructure story into a time series. It reports monthly and annual throughput, reliability, prorations and deferred barrels. The page says TAPS throughput peaked in 1988 at an average of 2,032,928 barrels per day, fell below 500,000 barrels per day in 2019, and reached an all-time low annual average of 462,821 barrels per day in 2025. For May 2026, it listed a daily average of 452,542 barrels per day, year-to-date daily average of 461,891 barrels per day and 100 percent reliability with no prorations or deferred barrels for the month.

Those numbers matter because declining throughput changes the physical behavior of the asset. Alyeska's flow-assurance page explains the issue in plain operational terms: less oil means slower-moving oil, slower-moving oil means colder oil, and colder, slower oil complicates the operator's work. Water can separate from crude and accumulate at low points, increasing corrosion and ice-related risk during shutdowns. Wax can deposit on pipe walls or settle out when the crude is cooler and less turbulent, requiring cleaning pigs and other mitigations.

This makes low flow a data problem. Alyeska needs timely temperature, flow, pressure, shutdown, pigging, corrosion and field-inspection evidence to decide when a mitigation is working. The public throughput table gives outside readers a monthly surface, but the operational surface has to be much finer. Operators need to understand changing conditions within the line, not only the annual average. Engineers need to know whether low-flow assumptions still fit the actual crude behavior and whether cleaning, heating, recirculation, pressure control or other mitigations are keeping risk inside acceptable boundaries.

The commercial question follows. The assignment asks whether storage, compute, migration, lock-in and data-quality labor beat the current stack. In Alyeska's case, that cannot mean a buyer comparing two SaaS dashboards. It means whether the cost of maintaining modernized control systems, data-gathering equipment, maintenance records, integrity databases, dashboards and response readiness is lower than the risk created by stale or fragmented records. The alternative to disciplined data work is not free operation.

It is more manual reconciliation, slower incident escalation, less confidence in low-flow mitigations and potentially higher maintenance or shutdown cost.

The public record supports a basic conclusion: low-flow operation requires continuous evidence, not just historical reputation. A pipeline that operated well at high throughput cannot assume the same data model, maintenance cadence or restart logic at lower throughput. Wax, water, ice, corrosion and temperature do not respond to brand history. They respond to physics and field conditions. Alyeska's public flow-assurance discussion is valuable because it acknowledges that lower throughput creates complicated challenges and that engineers are validating mitigations through laboratory and field tests.

The gap is that public readers cannot see the test data, acceptance thresholds or decision logs.

That gap is important because reliability percentages can be misunderstood. The public page's reliability metric is useful. Alyeska's 2025 sustainability report says TAPS reliability is calculated using deferred barrels caused by throughput disruption as a percentage of total throughput. That is a meaningful operating measure, but it is not the same as a full data-quality measure. A month with high throughput reliability does not prove every data lineage, maintenance record, alarm classification or cyber control worked perfectly.

It says the pipeline performed its intended transportation function without a measured deferred-barrel penalty under that metric.

For a technology reader, the best use of the throughput data is therefore not to declare victory or failure. It is to identify the live operating burden. Lower flow increases the need for fresh telemetry and disciplined maintenance records. Public reliability numbers show that Alyeska reports the transportation outcome. Flow-assurance materials show why the outcome is harder than it looks. The evidence limit is that the public cannot reproduce the internal calculations or inspect the supporting operational records.

Integrity data is where inspection becomes memory

Pipeline integrity is another place where Alyeska's operating record becomes the product surface. The 2024 Factbook describes cleaning pigs and instrumented, or smart, pigs. Cleaning pigs remove debris that could compromise sensor data. Smart pigs use nondestructive sensors to inspect for corrosion, deformation and movement. The Factbook describes ultrasonic transducer tools for wall thickness, magnetic flux leakage tools for metal loss, inertial measurement tools for pipe movement and curvature, and caliper tools for dents, ovalities, wrinkles and buckles.

Those details are not a list of gadgets. They describe a data pipeline inside the pipeline. A tool travels through the line, collects a stream of measurements, and produces evidence that engineers must locate, compare, prioritize and eventually turn into field work. If the cleaning run is poor, sensor data can be compromised. If location matching is weak, an anomaly can be assigned to the wrong spot. If historical inspection data is hard to compare with a new run, trend analysis weakens. If engineers cannot translate sensor output into excavation priorities, inspection becomes archive rather than action.

The public material also shows that integrity is not limited to corrosion. The 2025 sustainability report discusses work near the Jim River pingo, where an ice-cored hill and thaw settlement created possible pipeline instability and support challenges. Alyeska says teams used real-time mechanical modeling and field engineering to address the problem, retired thermal support members and heat pipes, and replaced them with non-thermal, deep-driven support structures in thaw-stable soils. The same report says the Integrity Management team monitors and responds to corrosion as well as civil integrity threats such as flooding and ground movement.

That is a strong subject-specific fact for this article's angle. Alyeska's critical-infrastructure data record has to account for climate, terrain and support behavior. A support issue is not a generic "asset maintenance" row. It is a geotechnical, environmental, structural and operational event that may interact with river movement, thawing permafrost, flood conditions and long-term reliability. A modernized database is useful only if it lets engineers connect these fields instead of leaving them in separate silos.

The sustainability report offers more measurable integrity and maintenance signals from 2025. Alyeska reported 16 excavations to mitigate corrosion and extend operational life of the mainline and fuel gas line, 12 vertical support member replacements, 65,000 cubic yards of rock hauled for river training and bank protection, and out-of-service inspections of two crude oil storage tanks. These are not private SCADA tests, but they are evidence that the public operating story includes proactive inspection, physical intervention and recordable maintenance.

The Valdez terminal adds a different integrity layer. Alyeska's terminal page says the facility includes process areas such as ballast water treatment and power generation, and that tanks are paired within containment dikes designed around the volume of both tanks with allowance for water and snow accumulation. The 2025 sustainability report describes the demolition of a long out-of-service ballast water storage tank that had become a safety concern because snow shedding could affect nearby people and in-service assets. The report says the work required careful planning around containment liners, active process areas, wildlife, waste, scrap handling and adjacent infrastructure.

That example illustrates why the operating record must be more than equipment inventory. A retired tank can still be a live risk if it sits near active assets. Demolition can reduce risk only if engineering packages, environmental protections, contractor coordination, waste records and safety controls are correct. The public evidence reports completion in about 90 days without environmental incidents, but it does not expose the underlying work package, permit history or daily field records. Again, public evidence establishes shape and outcome, not reproducible private performance.

For a buyer or regulator, the main integrity-data question is whether the record can survive repeated use. Can a smart-pig finding be traced to an excavation, a repair decision, a corrosion model update and a future inspection comparison? Can a vertical support member replacement be traced to geotechnical evidence and a new asset state? Can terminal demolition records prove that nearby containment and process risks were controlled? Alyeska's public sources suggest these workflows exist. The evidence available to outsiders cannot prove their internal latency, completeness or error rate.

Emergency readiness is a second operating system

Emergency preparedness is not an afterthought around Alyeska. It is a second operating system that sits beside the pipe. Alyeska's emergency preparedness page describes overlapping layers of protection: corrosion control, engineered design and review, preventive maintenance, condition monitoring, tank-level measurements, oil-flow measurements and volume-measurement systems. It says control and communication protections can shut down the system before a leak possibility is created, and that continuous monitoring by qualified individuals reduces both likelihood and consequence.

That language matters because it joins prevention and response. A response system that waits until a spill is visible is not enough. The data system has to detect abnormal conditions, support safe shutdown, route the event to people with authority and give field teams usable information. The public page says Alyeska prepares three publicly reviewed oil discharge prevention and contingency plans for the pipeline, the Valdez Marine Terminal and Prince William Sound, reviewed by state regulators. This creates a formal public-record surface around incident readiness.

The emergency reporting page adds the public interface. It identifies TAPS as an 800-mile, 48-inch oil pipeline plus a 148-mile underground fuel gas pipeline from Pump Station 1 to Pump Station 4 and other lines at the terminal and facilities. It gives the public a way to recognize possible leaks by smell, sight and sound, and it instructs callers to provide location, description and contact information from a safe place. That is not glamorous technology, but it is critical data collection at the edge. A public call is only useful if the response system can translate it into location, risk, dispatch and recordkeeping.

Prince William Sound adds a marine-response layer. Alyeska's SERVS page says the Ship Escort Response Vessel System was created in 1989 to prevent oil spills and provide spill-response and preparedness capabilities for Alyeska and the marine shipping companies calling at the Valdez Marine Terminal. It says SERVS works with the U.S. Coast Guard to monitor vessel traffic, that two tugs depart with every tanker, and that crews and equipment are staged around Prince William Sound. It lists response assets such as escort and response tugs, oil recovery barges, skimming units, boom, response centers, vessels of opportunity and on-water storage capacity.

From a data perspective, SERVS is a readiness database made physical. Vessels, crews, boom, barges, hatchery protection sites, response centers, weather, tanker movements and drills have to be known before an event. Alyeska says SERVS maintains a constant state of readiness and must have the equipment and personnel to recover 300,000 barrels of oil in 72 hours. That is a capability claim with operational implications. It depends on asset availability, crew qualification, logistics, locations, exercises and regulatory plan alignment. Public readers can see the published claim and asset categories.

They cannot test actual mobilization without an exercise record or incident record.

The 2025 sustainability report gives a useful public exercise signal. It says Alyeska's Emergency Preparedness and Response team maintains contingency plans and response equipment, provides training for incidents at the Valdez Marine Terminal, in Prince William Sound and along 800 miles of pipeline, and tracks compliance with federal and state prevention and response rules. It reports 139 field and Incident Management Team exercises in 2025, with 77 in Valdez and Prince William Sound and 62 along the pipeline. It also describes training academies, vessel training sessions and village response team training sessions.

These figures support a serious continuity posture, but they should not be oversold. Exercise counts do not reveal exercise quality, corrective-action closure, notification latency, contact-list freshness, equipment failure rates or after-action learning. They show repetition and public commitment. A technology assessment should ask what happens after the exercise: whether lessons update plans, whether field records update equipment inventories, whether people who missed the initial briefing receive the current control status and whether the response record is recoverable years later during audit, litigation or public review.

This is where public-sector continuity and data-sovereignty themes connect. Alyeska operates in Alaska, with Anchorage, Fairbanks and Valdez presence, a workforce that the company says is overwhelmingly Alaska-based, and a response system tied to local communities and sensitive marine areas. The data that supports this system is not a generic cloud artifact. It concerns Alaska's land, waters, workers, regulators, communities, tankers and public revenue. The public does not need every private record. It does need confidence that the records which govern local risk are maintained under accountable operating control.

Regulation and sustainability reporting reveal the governance surface

Alyeska's 2025 sustainability report is useful because it exposes governance signals that normal marketing pages do not. The report says Alyeska continued work on the Alyeska Management System, with continuous improvement, clearer roles and responsibilities, automated performance reporting, dashboards and improved reporting tools. It says the work aligned with API Recommended Practice 1173 for Pipeline Safety Management Systems and that an external API assessment noted progress with continuous improvement.

For a data-infrastructure reader, the words "automated performance reporting" and "dashboards" are less important than the governance function behind them. Dashboards can make weak records look authoritative if the inputs are stale or poorly defined. They can also make operating risk visible if the inputs are disciplined. The report's strongest claim is not that dashboards exist. It is that management-system work is being tied to roles, responsibilities, performance metrics, lessons learned and external assessment against a recognized pipeline safety management standard.

The report also gives a public view of regulatory intensity. It says TAPS is highly regulated and that Alyeska works with local, state and federal agencies daily. It identifies 18 agencies that oversee TAPS and reports hundreds of letters and requests for information received from regulators, hundreds of letters sent by Alyeska, 150 days in 2025 when agencies performed surveillance or inspection, and inspection outcomes that included satisfactory and unsatisfactory findings. Those numbers matter because they show that the operating record has to be usable by outsiders, not only internal managers.

Regulatory correspondence is a data workflow. A request arrives. The company must identify relevant records, check accuracy, respond, track commitments, preserve evidence and feed any required corrective action back into operating practice. If the underlying records are fragmented, every regulator request becomes manual archaeology. If records are governed well, the response can be faster and more consistent. The public report does not prove the latter. It does prove the volume and seriousness of the accountability surface.

The sustainability report also mentions cybersecurity, saying Alyeska's cybersecurity program helps the company innovate and manage cyber-related risks. That is an important acknowledgement, but it is not a public audit of cyber resilience. It does not disclose segmentation between enterprise and control systems, access-control rules, incident response metrics, vendor dependencies, backup design or penetration-test outcomes. A critical-infrastructure operator should not expose sensitive security details in a public article. The correct public conclusion is that cyber risk is acknowledged, while performance evidence remains private.

The Associated Press adds an independent pressure point. In 2024, AP reported that environmental groups petitioned the U.S. Department of the Interior to review climate impacts related to TAPS and consider a managed phasedown conversation. AP noted that the line began operating in 1977, that flow had fallen from about 2 million barrels per day at peak to roughly 470,000 barrels per day, and that the last major environmental review tied to right-of-way renewal was more than 20 years old.

The article also reported concerns about rapid Arctic warming, permafrost and future right-of-way analysis, while Alyeska said it continued to work with regulators on safe and environmentally responsible operations.

This external context is not a verdict against Alyeska. It is evidence that the operating record is politically and environmentally consequential. If a future review, renewal, challenge or phasedown debate accelerates, the quality of Alyeska's evidence will matter. Throughput history, low-flow mitigation, permafrost and flood response, integrity work, emergency exercises, terminal modernization and regulatory correspondence will all become part of the public argument. In critical infrastructure, data quality is not just an internal efficiency issue. It becomes institutional memory.

What the public evidence can and cannot establish

The public evidence can establish that Alyeska operates a real, mature and highly regulated infrastructure system. It can establish that TAPS is a long-distance, 48-inch pipeline with pump stations, terminal storage, response bases, communications systems, SCADA telemetry and smart-pig integrity workflows.

It can establish that monthly throughput and reliability are published, that low-flow physics are acknowledged as operationally challenging, that pump-station automation and remote control are part of the system, that emergency and spill-response plans exist, and that the company reports regulator engagement, inspections, exercises and management-system work.

The public evidence can also establish subject-specific technical dependency. Alyeska's technology surface is not a generic database. It is a record of pressure, flow, temperature, tank levels, valve positions, pump status, seismic monitoring, maintenance monitoring, corrosion inspections, pigging runs, terminal inventory, response equipment, regulatory correspondence and public reports. This is an operating record for Alaska critical infrastructure. It is directly tied to physical consequences.

What the public evidence cannot establish is equally important. It cannot prove SCADA uptime, alarm false-positive rates, cyber segmentation, identity and access management, backup and restore success, historian retention, data lineage tooling, mobile field application performance, incident-ticket latency, dashboard input quality, operator workload, control-room interface design or private recovery procedures. It cannot show whether a particular internal modernization project reduced failure rate or storage cost. It cannot benchmark query latency for an engineer trying to answer a regulator's question during an event.

That means no outside article should claim direct product testing. There is no public way to run an operational scenario through Alyeska's control room, submit a shipper workflow, inspect the response contact database, retrieve a smart-pig run, test a valve command, replay an incident drill or review cyber logs. The right language is conditional and evidence-based: public sources support the existence of a serious operating system, but they do not make the private control stack independently reproducible.

This distinction also prevents a common category error. Alyeska is not proven to be a "cloud service" just because the directory category uses a cloud-service taxonomy row or because modern operations require data systems. The public technology issue is operational data reliability, not a hosted customer application. The article's commercial question is therefore about the cost of reliable records versus the burden of legacy, fragmented or manually reconciled records. It is not a claim that Alyeska sells a storage platform to data teams.

The evidence limit also protects against unsupported reliability claims. Alyeska's 2025 annual reliability and May 2026 monthly reliability figures are useful, but they are not a full audit of every internal system. SERVS asset counts are useful, but they do not prove every deployment under real conditions. Exercise counts are useful, but they do not prove every after-action item closed. Cybersecurity acknowledgement is useful, but it is not an independent cyber review. The public record can support a serious evaluation only if these boundaries are respected.

The main failure modes are ordinary and severe

The assignment's known failure modes fit Alyeska unusually well: data staleness, monitoring gaps, integration breakage, incident escalation delay, OT/IT boundary ambiguity, continuity burden and unsupported reliability claims. None requires an exotic scenario. A stale asset status can mislead maintenance planning. A monitoring gap can delay recognition of abnormal operations. A broken integration can leave a field finding outside the central record. An unclear OT/IT boundary can increase cyber risk or slow safe data sharing. A weak continuity record can leave a regulator, responder or engineer reconstructing events manually.

Low-flow operation intensifies these ordinary risks. Slower and colder oil increases the importance of temperature, wax, water, pigging and corrosion evidence. If the record of cleaning pigs, instrumented inspection data and operating temperature trends is hard to query, mitigation decisions become more laborious. If public throughput stays high but field indicators are scattered, the reliability metric can look cleaner than the underlying work feels. That does not mean the system is unsafe. It means the data burden grows as the physical operating envelope changes.

Remote geography intensifies the same problem. Many assets are far from urban support. Weather, darkness, river movement, permafrost and access limitations can make the difference between a small intervention and a larger problem. A response plan has to know where equipment is staged and whether crews can reach it. A maintenance team has to know whether a support structure, relief station or standby system has changed status. A regulator request has to be answered with records that may involve multiple sites, years and contractors.

Terminal modernization adds another integration risk. The sustainability report says Alyeska made progress modernizing the Valdez Marine Terminal in 2025, including removal, inspection and legacy-system upgrades. It describes vapor-management improvements, a replacement vapor combustor with automatic ignition and remote startup capability, and routing vapors to boilers before sending remaining volumes to destruction.

These changes can improve safety, reliability and emissions performance, but they also create new handoff records: old equipment status, new equipment commissioning, control logic, maintenance procedures, emissions assumptions and operator training all have to be kept current.

Regulatory pressure adds a public-reachability risk. Existing published content and public records cannot be hidden to avoid hard questions. Alyeska's monthly throughput data, sustainability claims, emergency readiness, regulatory correspondence and public operating pages are part of the accountability surface. If a future review asks whether the line can operate safely under changing climate, lower flow or longer right-of-way horizons, the company will need evidence that can be assembled without excessive manual repair. A weak evidence chain can become a governance problem even before it becomes a physical failure.

The OT/IT boundary is especially important because Alyeska's public Factbook separates control data systems from enterprise data services, and the sustainability report acknowledges cybersecurity as a program area. Operational systems need protection from ordinary business-system risk, yet business, regulatory and engineering teams need enough data to make decisions. Too much isolation can create manual work and stale copies. Too much connectivity can increase security exposure. Public evidence does not show how Alyeska balances this. It only shows that the balance matters.

The most tempting unsupported claim would be that long operating history alone proves current data resilience. TAPS has operated since 1977, and that history is meaningful. But longevity can hide technical debt. Old assets accumulate exceptions, patches, retired stations, repurposed facilities, legacy naming conventions, changing ownership structures, different regulatory regimes and evolving climate assumptions. The more durable the infrastructure, the more important it is that the operating record be maintained as a living system rather than a historical archive.

How to judge Alyeska as a technology operator

The fair technology test for Alyeska is not whether it can make a dashboard. It is whether operational evidence remains usable when repeated decisions arrive under pressure. A control-room operator needs reliable telemetry. An integrity engineer needs inspection history that can be compared over time. A maintenance planner needs asset status that reflects field reality. An emergency manager needs equipment and people records that are current. A regulator needs traceable evidence. A public reader needs enough disclosure to understand the boundary between verified facts and private systems.

On the public record, Alyeska has several strengths. It publishes detailed throughput history and current monthly data. It discloses low-flow challenges rather than pretending the physics are static. It describes SCADA telemetry, remote control, microwave and fiber communications, seismic and maintenance monitoring, pump-station automation, smart-pig inspection, emergency plans, SERVS capabilities, response exercises and management-system work. It reports workforce, regulatory, integrity and terminal-modernization activity in the sustainability report. These are serious signals.

The public record also leaves important questions unanswered. How quickly do alarm classifications converge across systems? How often do smart-pig findings lead to record corrections? How frequently do dashboards reveal stale inputs? What are the access rules between enterprise and control data? How are field mobility, backup, historian retention and cyber response tested? What is the recovery process if a control data path fails during severe weather? How are lessons from 139 exercises turned into updated plans? These questions cannot be answered from public pages.

That does not make the public record weak. It makes it bounded. Critical infrastructure cannot be fully transparent without becoming less secure. But the public-facing record should be specific enough to avoid generic trust claims. Alyeska's best public evidence is specific: pump stations in service, throughput history, flow-assurance risks, SCADA telemetry categories, pigging data types, terminal storage capacity, response equipment categories, regulator engagement volumes and integrity work examples. The weakest public claims would be broad words like innovation, reliability or safety without supporting operational detail.

For data teams outside the pipeline industry, Alyeska offers a practical lesson. The hardest data problems often appear after the physical system is mature. New projects get attention; long-running operations get exceptions. A 1970s infrastructure system that is still operating in 2026 has to reconcile old design assumptions, modern automation, low-flow physics, climate exposure, cyber risk, regulatory review and public scrutiny. The record has to be robust enough to survive decades of changes in people, tools, standards and ownership.

For public-sector continuity, the lesson is sharper. Alaska's economy, marine environment, emergency planning and public revenue history have been tied to TAPS. Decisions about continuation, modernization, review or eventual transition require evidence that can be trusted by more than the operator. Alyeska does not need to expose private control systems to the world. It does need public evidence that is coherent enough for regulators, communities and readers to understand why the system matters, what it can prove and what remains uncertain.

The most defensible conclusion is therefore specific rather than absolute. Alyeska Pipeline Service Company should be understood as a critical-infrastructure operator whose technology relevance lies in the operating data record around TAPS. Public evidence supports the existence of a sophisticated operating, integrity and response system with real telemetry, automation, maintenance and regulatory surfaces. Public evidence does not prove private data freshness, governance quality, queryability or recoverability under stress.

The company's technology story is the gap between those two facts: a pipeline that can only remain credible if the records behind it keep working.