Characterisation & Planning
Establishing and recording exactly what is on the site before anything is touched - the structures, their condition, the materials they are made of, what runs beneath them and where every stream of arisings will eventually be allowed to go.
Last updated 2026-08-25
Typical duration
Typically 2-5 years for the initial campaign before significant dismantling starts, with survey, sampling and record updating then continuing alongside the works for the whole life of the programme.
What is Characterisation & Planning?
Characterisation is the survey stage, but on a decommissioning site it carries far more weight than a dilapidation survey on an ordinary building. The job is to establish, and then record, exactly what is there: the structures, their condition, what they are made of, what is inside them, what runs beneath them and what has happened to them across a working life that may run to half a century. Nothing on a decommissioning site moves until it has been characterised and the record exists. That single rule is what separates this work from ordinary demolition, and it is why a decommissioning programme spends years surveying before a machine is started.
Every information source is imperfect. Drawings were superseded and never reissued. Modifications were made and recorded on a sketch that has since been lost. Buildings were extended, sub-divided, re-clad and re-purposed. Plant was replaced in kind and the replacement was not the same shape. The people who knew the plant retired a decade ago. So characterisation is part records archaeology and part physical survey. The desk study establishes what should be there, the walkdown establishes what is, and the gap between the two is where the programme risk lives. Modern survey helps enormously - laser scanning, photogrammetry and remotely deployed cameras produce a measured record of spaces nobody wants to stand in - but somebody still has to interpret what comes back.
What comes out of characterisation is not a survey report. It is a set of decisions. The classification of every structure and every stream of material decides where the arisings can go, and where they can go decides how they are cut, how they are packaged, how they are handled and what the whole programme costs. Get that wrong at the front end and it does not surface as a small error. It surfaces years later as a stack of containers no disposal route will accept. The plan that follows sets the end state, the strategy for reaching it and the sequence, and it is reviewed and accepted by the site licensee under arrangements agreed with the regulator before any of it is acted on. Radiological matters, security arrangements and anything to do with nuclear material are controlled by the licensee and the regulator and sit outside the scope of this guide.
Compare the methods at a glance
When and why is Characterisation & Planning used?
Characterisation comes first because everything downstream depends on it, and it never really stops - it is repeated every time a wall comes down and reveals something the record did not mention. The reason it is taken to this depth is that a decommissioning programme is a waste logistics operation wearing a demolition jacket. The physical act of cutting a pipe or dropping a wall is not difficult; the industry has been doing both for a century. The difficulty is that every piece produced has to go somewhere, the routes available are few, and each route accepts material only against the waste acceptance criteria set by that route. A piece of steel that has been characterised, recorded and shown to meet those criteria is an asset with a destination. The identical piece of steel with no record is a liability that stays on the site. The commercial logic follows directly from that. Characterisation is cheap compared with almost anything that happens later - a full sampling and survey campaign costs a fraction of a single interim store - and the cost of every subsequent operation is fixed by the classification the characterisation produced. Programmes that invest properly at the front end spend the next twenty years executing a plan. Programmes that do not spend the next twenty years making discoveries, and each discovery stops the work, reopens the safety case, changes the disposal route and puts another year on the end date.
Types of Characterisation & Planning
Explore each method in depth - benefits, limitations, plant and quality control on its own page.
Desk study and record reconstruction
Pulling together drawings, modification records, commissioning data, operating history, incident records and photographs into a single account of what was built and what was done to it. It is slow, unglamorous and the cheapest characterisation available, because every question it answers is a question nobody has to answer with a person in a difficult space.
Non-intrusive walkdown and remote survey
Measured survey of the structures as they stand - laser scanning, photogrammetry, pole and crawler cameras, and remotely deployed platforms where the space is not accessible. It produces a geometrically accurate record without opening anything up, and on most sites it is the first time the plant has been measured properly since it was built.
Intrusive investigation and sampling
Coring, drilling and sampling of concrete, steel, insulation, coatings and linings to establish what the material actually is and how it has to be classified. Every intrusion is planned, permitted and made good, because opening a structure changes its condition and creates arisings that then need a destination of their own.
Ground, drainage and buried infrastructure survey
Boreholes, trial pits, geophysics, drain surveys and service tracing to establish what is under the slab. Old industrial sites are full of abandoned tanks, culverts, ducts and drains that appear on no drawing, and finding them now rather than with an excavator bucket is the whole point of the exercise.
Best suited for
- Any plant coming out of service, where the record has to be established before the strategy or the cost can be set
- Sites with a long operating history, multiple modifications and drawings that were never brought up to date
- Structures with large volumes nobody can enter, where remote survey is the only realistic way to build a record
- Programmes about to commit to a disposal strategy, where the classification of the arisings drives everything downstream
- Sites heading for a long quiescent period, where the record has to survive and stay usable for decades
Characterisation & Planning: step by step
- 1
Step 1: Reconstruct the history from the records
The first campaign happens in an archive rather than on site. Original design drawings, as-built information, modification and maintenance records, commissioning data, operating logs, photographs and any previous survey work are gathered, indexed and reconciled against each other. Contradictions are recorded rather than resolved by preference, because a contradiction between two records is a flag for physical investigation later. Where the site has people who worked the plant, structured interviews are worth more than any drawing, and they have a short shelf life. The output is a controlled baseline of what the site believes is there, with the confidence in each part of it stated openly.
- 2
Step 2: Walk the plant down and measure what is actually there
Surveyors go through the buildings room by room with the baseline in hand and record reality. Laser scanning and photogrammetry give a measured, dimensionally accurate record of every accessible space, which is worth far more than a sketch because it can be interrogated for years afterwards by people who will never visit. Structural condition is assessed at the same time - corrosion, spalling, water ingress, movement, previous repairs and any temporary propping that has quietly become permanent. Access for the survey itself is planned as a job of work, because a lot of the plant was built to be operated, not to be inspected.
- 3
Step 3: Survey remotely where the space cannot be entered
Plenty of the volume on an old plant is not somewhere anyone is going to walk. Ducts, cells, voids, sumps, culverts and tank interiors are surveyed with pole cameras, crawler robots, remotely operated vehicles and small unmanned aircraft, all of which can be recovered, cleaned or written off in a way people cannot. The deployment is engineered rather than improvised: how the device gets in, how it gets out, what happens if it fails halfway, and how it is decontaminated afterwards. Access into any area under radiological control is set by the licensee under arrangements agreed with the regulator and sits outside the scope of this guide.
- 4
Step 4: Take samples and investigate the structure
Where the visual survey cannot answer the question, material is taken and tested. Concrete is cored to establish strength, reinforcement arrangement, cover and the condition of the section through its depth. Steel is sampled for composition, because composition affects both structural assessment and the eventual classification of the arisings. Coatings, linings, insulation, gaskets and floor finishes are sampled for the conventional hazardous materials that any building of that age contains. Every sample is located precisely, logged, held under a chain of custody and tied back to the model, because a result with no location is a result that cannot be used.
- 5
Step 5: Characterise the ground, drains and buried services
Below the slab is where the surprises are. Boreholes, trial pits, geophysical survey, ground penetrating radar, drain tracing and camera surveys build a picture of the made ground, the foundations, the buried tanks, the abandoned culverts and the drainage systems that were extended six times and drawn twice. Groundwater is sampled and monitoring wells are installed, because the behaviour of the ground and the water in it will govern the remediation strategy twenty years from now. This work is cheap while the buildings are standing and expensive once the site is a live demolition area, so it is done early.
- 6
Step 6: Classify the material streams against the routes available
The survey data is turned into material streams, and each stream is tested on paper against the routes that could take it. This is the step that decides the whole programme. The waste acceptance criteria set by each disposal route determine the form, the size, the packaging and the records that route will accept, and those criteria then work backwards into how the material is cut, handled and stored. Streams with no available route are identified now, not later, because a stream with nowhere to go becomes an interim store, and an interim store is a building project with a design life measured in decades. Segregating clean material from the rest is where the largest cost reduction on any decommissioning programme is found.
- 7
Step 7: Fix the end state and choose the strategy
The end state is a statement of what the land will be when the work is finished - open land, a brownfield industrial platform, or a site retained for another use. It sounds like a policy question and it is really an engineering one, because the end state determines how far down the structures come, how much of the below-ground work is removed and how far the ground remediation goes. Alongside it sits the strategy question: continuous dismantling from now until the end state, or dismantling in stages with a long quiescent period in the middle. Care and maintenance is a deliberate engineering choice, not neglect, and it is chosen on the basis of the condition of the structures, the availability of disposal routes and the sequencing of the wider programme.
- 8
Step 8: Write the plan, the sequence and the temporary works strategy
The characterisation is finally written into a plan: the order the structures come down, the techniques used on each, the containment and ventilation each one needs, where the arisings go at every stage, and how the site infrastructure is kept working while the buildings that house it are demolished. The temporary works strategy is developed alongside it, because progressive dismantling changes load paths in structures that were never designed to be taken apart. The plan is reviewed and accepted by the site licensee under arrangements agreed with the regulator before work starts, and it is a living document - it is revised every time the site tells the survey something new, which on a plant of this age is frequently.
Plant & equipment
- Terrestrial laser scanners, photogrammetry equipment and the mobile access needed to reach every level of a tall plant
- Pole cameras, crawler robots, remotely operated vehicles and small unmanned aircraft for inaccessible spaces
- Diamond coring and drilling rigs for concrete sampling, with water management and made-good arrangements
- Cable and service detection equipment, ground penetrating radar and drain camera survey rigs
- Cable percussion, rotary and window sampling rigs for boreholes, trial pits and monitoring wells
- Sample handling, containment and transfer equipment, with logging and chain of custody systems
- Structural monitoring instrumentation - crack gauges, tilt sensors and targets on structures being assessed
- Asset register, model and document management systems capable of holding the record for decades
Quality control & testing
- Every survey tied to a permanent site control network, so records taken twenty years apart can still be compared
- Sample locations recorded against the model, with chain of custody maintained from the point of taking to the result
- Confidence and uncertainty stated for every part of the baseline, rather than implied by a clean drawing
- Contradictions between records flagged and carried forward as investigation points rather than quietly resolved
- Sampling density and coverage set by a documented strategy, reviewed before the campaign rather than after it
- Material classification checked against the waste acceptance criteria set by each candidate disposal route
- Independent review of the characterisation before it is used to set the strategy or the end state
- The characterisation record placed under formal document control, because it will be relied on for decades
Safety watchpoints
- Structural condition of buildings that have had no investment for years - corrosion, spalling, fragile roofs and unrecorded propping
- Conventional hazardous materials throughout - asbestos, lead paint, refractory linings, oils and legacy chemicals
- Confined space entry into ducts, sumps, cells, culverts and tanks for survey purposes, under a formal permit regime
- Working at height on degraded structures where the fixings and the substrate are both unknown quantities
- Buried and unrecorded services during trial pits and boreholes, including services that are still live
- Intrusive work that opens up a structure and changes its condition or its containment
- Lone and small-team working in remote parts of a large site with no operating population left in the buildings
- Radiological protection, dose management and security arrangements, all controlled by the licensee and the regulator and outside the scope of this guide
Common defects to hunt
- Survey taken without site control, so the record cannot be tied to anything and has to be repeated
- Sampling density set by budget rather than by strategy, leaving whole structures classified on a single result
- Characterisation done against the wrong criteria, so material is packaged for a route that will not accept it
- Buried tanks, culverts and drains found by an excavator bucket rather than by a survey
- Records gathered but never reconciled, so the project holds three versions of the truth and trusts none of them
- The 3D model built once and never updated, so it diverges from the site within a year of the works starting
- End state changed late, after the sequence, the disposal routes and the containment strategy were all set around the old one
- Corporate memory lost because the people who ran the plant were never interviewed while they were still available
How long does Characterisation & Planning take?
Typical duration: Typically 2-5 years for the initial campaign before significant dismantling starts, with survey, sampling and record updating then continuing alongside the works for the whole life of the programme..
Related processes
Next in build sequence
02 - Asset Care & Preparation