기본 콘텐츠로 건너뛰기

추천 가젯

[로보틱스-2] 관절에 돈이 고인다

2026년 올해, 중국 저장에 로보틱스를 배우러 탐방을 간 적이 있다. 중국의 로봇굴기를 현지에서 느끼고자 나섰던 기회였다. 미래도시까지는 아니더라도 그들의 로봇에 대한 애정은 곳곳에 느껴졌다. 공산주의 정부의 강력한 로드맵은 어쩌면 시장주의보다 자원의 효율적 배분을 더욱 잘 이끌어 내는지도 모르겠다. 그래서, 중국의 굴기에 감탄과 공포에만 젖어 있어야 하는가? 나는 이 지점이 매우 불편했다. 한국의 AI 정책이 피부로 와닿는 지점은 결국 로봇이 될 텐데 어떻게 대응을 해야 할지 아무도 모르는 것만 같다는 어떠한 감을 느끼게 된다. 우리는 아마 매우 익숙한 경험이 있기 때문에 몸을 사리는지도 모른다. '수소'라는 아젠다가 공론이 되어 왔고 그 과정이 얼마나 무기력했는지도 알고 있다. 아무리 좋아 보여도 혼자 나대면 죽는다는 것을 한국의 직장인들은 너무나 잘 알고 있다. 다시 본론으로 돌아오자. 로봇 팔 하나를 열어 보면, 매끈한 흰색 껍데기 안 관절마다 손바닥만 한 쇳덩이가 박혀 있다. 감속기다. 빠르게 도는 모터의 힘을 느리고 세게 바꿔 주는 부품. 로봇이 무거운 것을 들고도 떨지 않는 건 대개 이 쇳덩이 덕분이다. 그리고 로봇 한 대 값의 절반쯤이 이 관절 언저리에서 갈린다. 사람들은 로봇의 얼굴을 보지만, 실제로 돈은 이 관절 역할을 하는 감속기에 모인다. 1부에서 나는 로봇의 머리 즉 두뇌 이야기를 했다. 지능이 담장 안으로 내려오고, 몸은 중국이 찍어내고, 머리는 미국에서 온다는 이야기. 이번에는 한 층 아래로 내려간다. 손목과 몸통. 실제로 쇠와 톱니로 만들어지고, 실제로 돈이 오가는 자리다. 이건 결국 중국과의 원가 싸움 아닌가 본론으로 내려가기 전에, 정직한 질문 하나를 통과해야 한다. 로봇은 결국 제조업이다. 쇠를 깎고, 감고, 조이고, 상자에 담아 파는 일. 그리고 제조업의 원가 싸움에서 중국을 이겨 본 기억을, 나는 최근 20년 사이에 몇 개 떠올리지 못한다. 태양광과 LCD가 어떻게 끝났는지 우리는 안다. 그렇다면 한국 로보...

[Value Chain Journey-4] The People Who Fix Things — Power Plant Maintenance

One night last summer, the breaker in my home tripped and the whole place went dark. Feeling my way to the fuse box by flashlight, I thought: surely I am not the only person who thinks about electricity only after it is gone. A power plant, once it starts turning, does not readily stop. Because stopping means disaster. When one large generator goes down without warning, billions of won vanish in a single day; when several go down at once, a city's lights waver. So attached to every plant are people who watch over the vast machinery night and day, so that it never stops for even a moment — replacing worn parts, putting hands on trouble before it becomes a breakdown. The work is called power plant maintenance. We barely know these people exist. But without them, the expensive transformers of the last chapter, and the plants themselves, are so much scrap iron.

This journey goes to those invisible people. And here, for the first time in this book, I find myself looking long at one person's name.

The Door a Public Company Held Has Opened

Power plant maintenance was, for a long time, the de facto monopoly of a public company called KEPCO KPS. The plants were mostly built by the state, so it seemed natural that the state's company should fix them. Like the Korea Exchange or the telecoms we saw earlier, a solid wall stood here too. But a door opened in this wall as well, and this time the private company that walked through it succeeded in pushing the wall back. Soosan Industries. In thermal power maintenance, the public company's share is being pushed back year after year from around half, and at the very front of the private companies pressing into the opening stands this one. And in the hardest territory of all: close to half this company's revenue comes from maintaining nuclear power plants — work not just anyone is allowed to touch.

A Wall Crossed Only by Time

Why can't just anyone do it? Because putting your hands on a running reactor and turbine calls for something money cannot buy. Skill layered up over decades of taking the same equipment apart and putting it back together. The collective memory of people who know, when a certain part makes a certain sound, what to suspect. I have watched the mechanic at a long-standing garage listen to an engine turn over and point straight to the trouble; this, so to speak, is the reactor edition of that. It is precisely the same material as the "wall built of time" we saw at Samsung Electronics and SK hynix. Only there, an enormous conglomerate built the wall — and here, a maintenance company no one was watching built one the same way, and broke through a public company's fortress.

A Man Standing at the Source

When I read the life of the man who founded this company, something in me goes quiet. Jung Suk-hyun. Born in 1952 in Jangsu, North Jeolla Province, the youngest of five children — four daughters and one son. He entered middle school at the top of the entrance list and attended on scholarship, but the family's circumstances put college out of the question. He finished a technical high school in Jeonju, joined Hyundai E&C through its open recruitment for high school graduates, and worked by day while attending university at night. What he did on power plant construction sites looks unremarkable: he kept detailed records of the work. Only he did not let the records simply pile up. After each project was completed he analyzed them, searching for how to cut cost and schedule and raise safety. After ten years of this he handed in his resignation, started out in 1980 as a small tool dealer, and in 1983 founded the company that stands today.

And decades later, this company won the maintenance contract for the Barakah nuclear plant in the United Arab Emirates — alone, over global giants like General Electric and Siemens. The hands of a technical-school graduate born with nothing, who attended university at night and kept records on the construction site, had reached the place where the world's most exacting nuclear plants are repaired.

I have seen many splendid walls in this book. Walls built by law, walls built by habit, walls built by time. Some of them had things to apologize for; some we could tear down with our own hands. But this wall is of a different grain. It was raised by holding on to the unregarded work of fixing things — not glamorous semiconductors, not platforms, but the lowest place of all — with nothing but time, records, and stubborn persistence. In the preface to this series I wrote that one should stand where the deepest source is. The name Jung Suk-hyun shows what that sentence looks like when it takes the shape of a person.

The People Who Don't Leave — Here, the Business Itself

Throughout this book I have counted, company by company, "the people who don't leave" — because I took tenure as testimony to how solid a wall is. But in power plant maintenance, the counting goes beyond metaphor. Here, people are the equipment; one skilled hand who has stayed long is, in himself, this company's asset. Maintenance is done not by machines but by hands and memory, so when a person leaves, the wall grows lower with him. At this company, then, "people stay long" is simply another name for competitiveness.

CompanyEmployeesAverage payAverage tenureRevenue per employeeOperating profit per employee
Soosan Industries1,214about 70.7 million won8.5 yearsabout 180 million wonabout 22 million won
KEPCO KPS6,485about 83.2 million won13.9 yearsabout 240 million wonabout 22 million won

Based on FY2025 disclosures. Soosan Industries figures are on a separate financial statement basis (1,041 of its 1,214 employees are technical staff); for KEPCO KPS, consolidated and separate figures are nearly identical.

I set these numbers beside the chains that came before. Revenue per employee at the semiconductor equipment company: 780 million won. At the power equipment makers: 1.5 to 1.8 billion. In maintenance: 180 million won. Does it look shabby? I think this lowness is precisely the identity of the industry. Maintenance sells not equipment but people's time, so revenue is labor cost and the person is the product. That the company keeps only some 20 million won per head is for the same reason. And a small per-person number also means that the person cannot be subtracted from the work. If machines could have replaced them, the number would have grown long ago. Where value does not pool, people pool instead.

While the plants turn twenty-four hours a day, we do not think of the people standing behind them — the way we take it for granted that the light comes on when we press the switch. But what holds up that taken-for-grantedness is a pair of hands, somewhere in a plant that never goes dark, listening for a sound and changing a part. The work that looks lowest turns out, because no one can take its place, to be the highest wall. Tracing my way up the chain, I stayed at this station a long time. Because here I found myself looking past where the value pools, to where the person who made that value came from.

The next journey turns to the work of putting the electricity that plant makes into a vessel and keeping it there. A chain as difficult as making electricity: storing it.

This piece is part of the "Korea Value Chain" series and is an original work. It is an analysis based on public data and industry sources; some figures and assessments are estimates. It is not intended as a direct basis for investment decisions.

By Alexandro Lee · July 2026

댓글

가장 많이 본 글