Media room · 2026

Explain the part of the factory story that its headline number left out.

Industrial Data Bearings

Invented editorial series examining the meaning of industrial records

Eighty-six codes concealed 112 physical lots

I would revisit the count I endorsed in 2008 and the twenty-six codes reused across two factories. The proposed article would retain Mei Arvellan’s objection, S$640,000 review expense and withdrawal of my complete-traceability claim. It would distinguish the correction of a data interpretation from any product-quality or safety conclusion, showing why responsibility includes reaching earlier readers rather than quietly replacing a number in a later report.

Industrial Data Bearings

Invented editorial series examining the meaning of industrial records

The old part number still explained an earlier record

I would examine why 2100 aliases and 2000 superseded entries remained meaningful beside 8300 current identifiers in the part-master review. The proposed account would preserve Darin Veylun’s preference for a shorter catalogue and S$3.7 million review expense. It would separate a convenient current view from the relationships needed to interpret history, without implying that a data team can establish physical interchangeability or claim an inventory saving from deleting records.

Industrial Data Bearings

Invented editorial series examining the meaning of industrial records

The arrival time was not the time the machine event occurred

I would describe the thirty-seven offset sources found within two hundred machine logs, retaining the four-to-eleven-minute range and S$1.9 million review expense. The proposed piece would explain why original timestamps and documented interpretation both remained necessary. It would not supply equipment-setting instructions or infer causation from a revised sequence. The editorial question concerns the moment represented by a field before that field supports a larger operational account.

Industrial Data Bearings

Invented editorial series examining the meaning of industrial records

An unfamiliar-machine claim still contained familiar assets

I would examine the forty-two-asset maintenance study through the question its evaluation could actually answer. A random split of observations left familiar asset context on both sides, while the intended claim concerned an unfamiliar machine. The proposed article retains S$2.3 million further-analysis expense and the withheld claim. It invents neither a later accuracy score nor a maintenance result, and leaves model evaluation and operating decisions with their responsible specialists.

Industrial Data Bearings

Invented editorial series examining the meaning of industrial records

Sixty-seven absent supplier values were not sixty-seven zeros

I would distinguish the 144 received measurements, fifty-nine modelled estimates and sixty-seven missing values within the 270-record supplier review. The proposed account keeps S$4.6 million expense and the inconvenience of an incomplete summary visible. It would explain why a received measurement is not automatically verified, without claiming assurance, a footprint reduction or a formal reporting conclusion. The story concerns the evidence behind a total that initially looked more complete than its records allowed.

Media biography

Surya Koh is an invented Singaporean Chief Data Officer whose illustrative career follows factory analysis into enterprise data responsibility. He joined the fictional Tervanex Components in August 1990 and became CDO in March 2019. His current function includes 158 colleagues through seven direct reports and S$21.4 million in delegated 2026 operating expense. Company scale, engineering authority and plant responsibilities are stated separately.

His fictional record examines the physical meaning behind part identifiers, production lots and machine timestamps, alongside the scope of model evaluation and supplier evidence. An owned lot-count error remains visible, including its correction expense and the claim withdrawn. He has executive board-presentation exposure without a statutory directorship.

Industrial Data Bearings is invented. Its listed angles are proposals rather than published coverage, and no external endorsement or speaking invitation is claimed. Every employer, educational institution, colleague and result in the demonstration is fictional. A real media account would need verified appointments, authorised examples and accurate publication links. This profile illustrates a way to explain industrial data judgement without borrowing engineering credentials or inventing operational outcomes.

Biographies

Approved text

Short speaker biography

Surya Koh is a fictional Singaporean Chief Data Officer at Tervanex Components. His illustrative manufacturing career examines the physical events behind factory codes, timestamps and analytical records. He leads 158 data colleagues through seven direct reports. His proposed discussions retain an owned lot-count error and distinguish data interpretation from engineering, quality, safety and reporting authority.

Standard speaker biography

Surya Koh is an invented industrial data leader whose career at the fictional Tervanex Components began in August 1990. He progressed from factory analysis through master-data supervision and plant analytics to enterprise data leadership, becoming CDO in March 2019. His current remit includes 158 colleagues, seven direct reports and S$21.4 million in delegated 2026 function expense. The illustrative cases examine part histories, machine timestamps, lot identities, model evaluation and supplier evidence. An owned 2008 counting error remains in the record. He has executive board-presentation exposure, without a statutory appointment. All employers, education, colleagues and results are fictional. The proposed sessions demonstrate how a leader could discuss authorised industrial evidence without claiming delivered engagements or directing plant operations.

Extended speaker biography

Surya Koh is a fictional Chief Data Officer based in Singapore. His invented career concerns the relationship between an industrial record and the physical event it is used to describe. After completing industrial statistics at Neraveth Institute of Industrial Statistics in June 1990, he joined Tervanex Components that August. Factory analysis led to master-data supervision, plant analytics and enterprise data responsibility before his appointment as CDO in March 2019.

He leads 158 data colleagues through seven direct reports, with S$21.4 million in delegated 2026 operating expense. The company has six factories and 6700 employees in this demonstration. Those company measures are distinct from his personal authority. Engineering, operations, quality, safety, IT/OT security and finance retain separate responsibilities.

His proposed talks follow five specific cases, including his incorrect inference that eighty-six codes represented eighty-six physical lots. Other examples address retained part histories, timestamp meaning, evaluation on unfamiliar assets and missing supplier evidence. He has presented to the board as an executive but has no statutory directorship in the fictional record. Every employer, colleague, quotation and result is invented. A real speaker account would need verified appointments, permission to discuss industrial records and a clear separation between proposed sessions and completed engagements. The purpose is to make data leadership understandable through decisions and corrections rather than decorative technical claims.

An invitation

Contact

I welcome a well-defined discussion about the evidence behind an industrial data proposition, or a first-board opportunity where this experience fills a specific gap. The mandate, preparation and employer position would need examination. This is an invented profile with no live advisory service. A real leader’s version should offer an authorised introduction route and keep plant-sensitive, customer and supplier material out of public enquiries.

Open to

Board appointments
Speaking invitations

Write to Surya

office@suryakoh.example