Calculate Embodied Carbon (co2) for Revit/ifc Models Using AI Classification

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "id": "UgJXCTdg8rea9Skt",
  "name": "n8n_7_Carbon_Footprint_CO2_Estimator_for_Revit and_IFC",
  "tags": [],
  "nodes": [
    {
      "id": "86512d4b-52b7-46ac-ac59-61ed748b3045",
      "name": "Find Category Fields1",
      "type": "n8n-nodes-base.code",
      "position": [
        -80,
        720
      ],
      "parameters": {
        "jsCode": "const items = $input.all();\nif (items.length === 0) {\n  return [{json: {error: 'No grouped data found'}}];\n}\n\nconst headers = Object.keys(items[0].json);\n\nconst categoryPatterns = [\n  { pattern: /^category$/i, type: 'Category' },\n  { pattern: /^ifc[\\s_-]?type$/i, type: 'IFC' },\n  { pattern: /^host[\\s_-]?category$/i, type: 'Host' },\n  { pattern: /^ifc[\\s_-]?export[\\s_-]?as$/i, type: 'Export' },\n  { pattern: /^layer$/i, type: 'Layer' }\n];\n\nlet categoryField = null;\nlet categoryFieldType = 'None';\n\n// \u0418\u0449\u0435\u043c \u043f\u043e\u043b\u0435 \u043a\u0430\u0442\u0435\u0433\u043e\u0440\u0438\u0438\nfor (const header of headers) {\n  for (const {pattern, type} of categoryPatterns) {\n    if (pattern.test(header)) {\n      categoryField = header;\n      categoryFieldType = type;\n      break;\n    }\n  }\n  if (categoryField) break;\n}\n\nconst volumetricPatterns = /volume|area|length|count|quantity|thickness|perimeter|depth|size|dimension|weight|mass/i;\nconst volumetricFields = headers.filter(header => volumetricPatterns.test(header));\n\nconst categoryValues = new Set();\nif (categoryField) {\n  items.forEach(item => {\n    const value = item.json[categoryField];\n    if (value && value !== '' && value !== null) {\n      categoryValues.add(value);\n    }\n  });\n}\n\nconsole.log('Category field analysis:');\nconsole.log('- Field found:', categoryField || 'None');\nconsole.log('- Field type:', categoryFieldType);\nconsole.log('- Unique values:', categoryValues.size);\nconsole.log('- Volumetric fields:', volumetricFields.length);\n\nreturn [{\n  json: {\n    categoryField: categoryField,\n    categoryFieldType: categoryFieldType,\n    categoryValues: Array.from(categoryValues),\n    volumetricFields: volumetricFields,\n    groupedData: items.map(item => item.json),\n    totalGroups: items.length\n  }\n}];"
      },
      "typeVersion": 2
    },
    {
      "id": "ceeaa6f3-781e-421f-bae1-c6ccb784ecbd",
      "name": "Apply Classification to Groups1",
      "type": "n8n-nodes-base.code",
      "position": [
        448,
        768
      ],
      "parameters": {
        "jsCode": "const categoryInfo = $node['Find Category Fields1'].json;\nconst groupedData = categoryInfo.groupedData;\nconst categoryField = categoryInfo.categoryField;\nconst volumetricFields = categoryInfo.volumetricFields || [];\n\nlet classifications = {};\nlet buildingCategories = [];\nlet drawingCategories = [];\n\ntry {\n  const aiResponse = $input.first().json;\n  const content = aiResponse.content || aiResponse.message || aiResponse.response || '';\n  \n  const jsonMatch = content.match(/\\{[\\s\\S]*\\}/);\n  if (jsonMatch) {\n    const parsed = JSON.parse(jsonMatch[0]);\n    classifications = parsed.classifications || {};\n    buildingCategories = parsed.building_categories || [];\n    drawingCategories = parsed.drawing_categories || [];\n    console.log(`AI classified ${Object.keys(classifications).length} categories`);\n    console.log(`Building categories: ${buildingCategories.length}`);\n    console.log(`Drawing categories: ${drawingCategories.length}`);\n  }\n} catch (error) {\n  console.error('Error parsing AI classification:', error.message);\n}\n\nreturn groupedData.map(group => {\n  let isBuildingElement = false;\n  let reason = '';\n  let confidence = 0;\n  \n  if (categoryField && group[categoryField]) {\n    const categoryValue = group[categoryField];\n    \n    if (classifications[categoryValue] !== undefined) {\n      isBuildingElement = classifications[categoryValue];\n      confidence = 95;\n      reason = `Category '${categoryValue}' classified by AI as ${isBuildingElement ? 'building element' : 'drawing/annotation'}`;\n    } else {\n      \n      const lowerCategory = categoryValue.toLowerCase();\n      const drawingKeywords = /annotation|drawing|text|dimension|tag|view|sheet|grid|section|elevation|callout|revision|legend|symbol|mark|note|detail items|filled region|detail line/i;\n      const buildingKeywords = /wall|floor|roof|column|beam|door|window|stair|pipe|duct|equipment|fixture|furniture/i;\n      \n      if (drawingKeywords.test(lowerCategory)) {\n        isBuildingElement = false;\n        confidence = 85;\n        reason = `Category '${categoryValue}' matched drawing keywords`;\n      } else if (buildingKeywords.test(lowerCategory)) {\n        isBuildingElement = true;\n        confidence = 85;\n        reason = `Category '${categoryValue}' matched building keywords`;\n      } else {\n        // \u041f\u043e \u0443\u043c\u043e\u043b\u0447\u0430\u043d\u0438\u044e \u0441\u0447\u0438\u0442\u0430\u0435\u043c \u0441\u0442\u0440\u043e\u0438\u0442\u0435\u043b\u044c\u043d\u044b\u043c \u044d\u043b\u0435\u043c\u0435\u043d\u0442\u043e\u043c \u0435\u0441\u043b\u0438 \u043d\u0435 \u043e\u0447\u0435\u0432\u0438\u0434\u043d\u043e \u043e\u0431\u0440\u0430\u0442\u043d\u043e\u0435\n        isBuildingElement = true;\n        confidence = 70;\n        reason = `Category '${categoryValue}' assumed as building element (no clear match)`;\n      }\n    }\n  } else {\n    // \u0415\u0441\u043b\u0438 \u043d\u0435\u0442 \u043a\u0430\u0442\u0435\u0433\u043e\u0440\u0438\u0438, \u043f\u0440\u043e\u0432\u0435\u0440\u044f\u0435\u043c \u043d\u0430\u043b\u0438\u0447\u0438\u0435 \u043e\u0431\u044a\u0435\u043c\u043d\u044b\u0445 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u043e\u0432\n    let hasSignificantVolumetricData = false;\n    let volumetricCount = 0;\n    \n    for (const field of volumetricFields) {\n      const value = parseFloat(group[field]);\n      if (!isNaN(value) && value > 0) {\n        hasSignificantVolumetricData = true;\n        volumetricCount++;\n      }\n    }\n    \n    isBuildingElement = hasSignificantVolumetricData;\n    confidence = hasSignificantVolumetricData ? 80 : 60;\n    reason = hasSignificantVolumetricData ? \n      `Has ${volumetricCount} volumetric parameters with values` : \n      'No category field and no significant volumetric data';\n  }\n  \n  return {\n    json: {\n      ...group,\n      is_building_element: isBuildingElement,\n      element_confidence: confidence,\n      element_reason: reason\n    }\n  };\n});"
      },
      "typeVersion": 2
    },
    {
      "id": "ffc937c8-4ca6-4963-a24f-114f45b5345b",
      "name": "Non-Building Elements Output1",
      "type": "n8n-nodes-base.set",
      "position": [
        848,
        784
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "message",
              "name": "message",
              "type": "string",
              "value": "Non-building elements (drawings, annotations, etc.)"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "731a2f76-60b4-4b4c-8e56-465c0ea6ad5c",
      "name": "Extract Headers and Data1",
      "type": "n8n-nodes-base.code",
      "position": [
        80,
        496
      ],
      "parameters": {
        "jsCode": " const items = $input.all();\n if (items.length === 0) {\n  throw new Error('No data found in Excel file');\n }\n\n const allHeaders = new Set();\n items.forEach(item => {\n  Object.keys(item.json).forEach(key => allHeaders.add(key));\n });\n\n const headers = Array.from(allHeaders);\n const cleanedHeaders = headers.map(header => {\n  return header.replace(/:\\s*(string|double|int|float|boolean|number)\\s*$/i, '').trim();\n });\n\n const headerMapping = {};\n headers.forEach((oldHeader, index) => {\n  headerMapping[oldHeader] = cleanedHeaders[index];\n });\n\n const sampleValues = {};\n cleanedHeaders.forEach((header, index) => {\n  const originalHeader = headers[index];\n  for (const item of items) {\n    const value = item.json[originalHeader];\n    if (value !== null && value !== undefined && value !== '') {\n      sampleValues[header] = value;\n      break;\n    }\n  }\n  if (!sampleValues[header]) {\n    sampleValues[header] = null;\n  }\n });\n\n console.log(`Found ${headers.length} unique headers across ${items.length} items`);\n\n return [{\n  json: {\n    headers: cleanedHeaders,\n    originalHeaders: headers,\n    headerMapping: headerMapping,\n    sampleValues: sampleValues,\n    totalRows: items.length,\n    totalHeaders: headers.length,\n    rawData: items.map(item => item.json)\n  }\n }];"
      },
      "typeVersion": 2
    },
    {
      "id": "7bbf3d9d-4406-4fa3-9ca6-49eeaea551ff",
      "name": "AI Analyze All Headers1",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "position": [
        240,
        496
      ],
      "parameters": {
        "modelId": {
          "__rl": true,
          "mode": "list",
          "value": "chatgpt-4o-latest",
          "cachedResultName": "CHATGPT-4O-LATEST"
        },
        "options": {
          "temperature": 0.1
        },
        "messages": {
          "values": [
            {
              "role": "system",
              "content": "You are an expert in construction classification systems. Analyze building element groups and assign aggregation methods for grouping data.\n\nRules:\n1. 'sum' - for quantities that should be totaled:\n   - Volume, Area, Length, Width, Height, Depth, Size\n   - Count, Quantity, Number, Amount, Total\n   - Thickness, Perimeter, Dimension\n   - Weight, Mass, Load\n   - Any measurable physical property that accumulates\n\n2. 'mean' (average) - for rates and unit values:\n   - Price, Cost, Rate (per unit)\n   - Coefficient, Factor, Ratio\n   - Percentage, Percent\n   - Efficiency, Performance metrics\n   - Any per-unit or normalized values\n\n3. 'first' - for descriptive/categorical data:\n   - ID, Code, Number (when used as identifier)\n   - Name, Title, Description\n   - Type, Category, Class, Group\n   - Material, Component, Element\n   - Project, Building, Location\n   - Status, Phase, Stage\n   - Any text or categorical field\n\nIMPORTANT: \n- Analyze each header carefully\n- Consider both the header name AND sample value\n- Return aggregation rule for EVERY header provided\n- Use exact header names from input\n\nReturn ONLY valid JSON in this exact format:\n{\n  \"aggregation_rules\": {\n    \"Header1\": \"sum\",\n    \"Header2\": \"first\",\n    \"Header3\": \"mean\"\n  }\n}"
            },
            {
              "content": "Analyze these {{ $json.totalHeaders }} headers and determine aggregation method for each:\n\nHeaders with sample values:\n{{ JSON.stringify($json.sampleValues, null, 2) }}\n\nProvide aggregation rule for EACH header listed above."
            }
          ]
        }
      },
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.3
    },
    {
      "id": "10caf6f7-ad6b-4de2-9e63-8b06b5609414",
      "name": "Process AI Response1",
      "type": "n8n-nodes-base.code",
      "position": [
        592,
        496
      ],
      "parameters": {
        "jsCode": "// \u041e\u0431\u0440\u0430\u0431\u0430\u0442\u044b\u0432\u0430\u0435\u043c \u043e\u0442\u0432\u0435\u0442 AI \u0438 \u043f\u0440\u0438\u043c\u0435\u043d\u044f\u0435\u043c \u043f\u0440\u0430\u0432\u0438\u043b\u0430\n const aiResponse = $input.first().json;\n const headerData = $node['Extract Headers and Data1'].json;\n\n// \u0418\u0437\u0432\u043b\u0435\u043a\u0430\u0435\u043c \u043f\u0440\u0430\u0432\u0438\u043b\u0430 \u0438\u0437 AI \u043e\u0442\u0432\u0435\u0442\u0430\n let aiRules = {};\n try {\n  const content = aiResponse.content || aiResponse.message || aiResponse.response || '';\n  console.log('AI Response received, length:', content.length);\n  \n  // \u0418\u0449\u0435\u043c JSON \u0432 \u043e\u0442\u0432\u0435\u0442\u0435\n  const jsonMatch = content.match(/\\{[\\s\\S]*\\}/);\n  if (jsonMatch) {\n    const parsed = JSON.parse(jsonMatch[0]);\n    aiRules = parsed.aggregation_rules || parsed.parameter_aggregation || {};\n    console.log(`AI provided ${Object.keys(aiRules).length} rules`);\n  } else {\n    console.warn('No JSON found in AI response');\n  }\n } catch (error) {\n  console.error('Error parsing AI response:', error.message);\n }\n\n// \u0421\u043e\u0437\u0434\u0430\u0435\u043c \u0444\u0438\u043d\u0430\u043b\u044c\u043d\u044b\u0435 \u043f\u0440\u0430\u0432\u0438\u043b\u0430 \u0441 \u0434\u0435\u0444\u043e\u043b\u0442\u0430\u043c\u0438\n const finalRules = {};\n headerData.headers.forEach(header => {\n  if (aiRules[header]) {\n    finalRules[header] = aiRules[header];\n  } else {\n    // \u041f\u0440\u0438\u043c\u0435\u043d\u044f\u0435\u043c \u043f\u0440\u0430\u0432\u0438\u043b\u0430 \u043f\u043e \u0443\u043c\u043e\u043b\u0447\u0430\u043d\u0438\u044e\n    const lowerHeader = header.toLowerCase();\n    \n    if (lowerHeader.match(/volume|area|length|width|height|count|quantity|thickness|perimeter|depth|size|dimension|weight|mass|total|amount|number/)) {\n      finalRules[header] = 'sum';\n    } else if (lowerHeader.match(/price|rate|cost|coefficient|factor|percent|ratio|efficiency|avg|average|mean/)) {\n      finalRules[header] = 'mean';\n    } else {\n      finalRules[header] = 'first';\n    }\n  }\n });\n\n const groupByParam = $node['Setup - Define file paths'].json.group_by;\n\n console.log(`\\nAggregation rules summary:`);\n console.log(`- Total headers: ${headerData.headers.length}`);\n console.log(`- AI rules: ${Object.keys(aiRules).length}`);\n console.log(`- Default rules: ${headerData.headers.length - Object.keys(aiRules).length}`);\n console.log(`- Group by: ${groupByParam}`);\n\n// \u0412\u043e\u0437\u0432\u0440\u0430\u0449\u0430\u0435\u043c \u0432\u0441\u0435 \u0434\u0430\u043d\u043d\u044b\u0435 \u0434\u043b\u044f \u0433\u0440\u0443\u043f\u043f\u0438\u0440\u043e\u0432\u043a\u0438\n return [{\n  json: {\n    aggregationRules: finalRules,\n    headerMapping: headerData.headerMapping,\n    headers: headerData.headers,\n    originalHeaders: headerData.originalHeaders,\n    rawData: headerData.rawData,\n    groupByParam: groupByParam,\n    totalRows: headerData.totalRows\n  }\n }];"
      },
      "typeVersion": 2
    },
    {
      "id": "435f2cb9-9ccb-4f32-ba95-6bb0126a73d2",
      "name": "Group Data with AI Rules1",
      "type": "n8n-nodes-base.code",
      "position": [
        784,
        496
      ],
      "parameters": {
        "jsCode": " const input = $input.first().json;\n const aggregationRules = input.aggregationRules;\n const headerMapping = input.headerMapping;\n const rawData = input.rawData;\n const groupByParamOriginal = input.groupByParam;\n\n const groupByParam = headerMapping[groupByParamOriginal] || groupByParamOriginal;\n\n console.log(`Grouping ${rawData.length} items by: ${groupByParam}`);\n\n const cleanedData = rawData.map(item => {\n  const cleaned = {};\n  Object.entries(item).forEach(([key, value]) => {\n    const newKey = headerMapping[key] || key;\n    cleaned[newKey] = value;\n  });\n  return cleaned;\n });\n\n const grouped = {};\n\n cleanedData.forEach(item => {\n  const groupKey = item[groupByParam];\n  \n  if (!groupKey || groupKey === '' || groupKey === null) return;\n  \n  if (!grouped[groupKey]) {\n    grouped[groupKey] = {\n      _count: 0,\n      _values: {}\n    };\n    \n    Object.keys(aggregationRules).forEach(param => {\n      if (param !== groupByParam) {\n        grouped[groupKey]._values[param] = [];\n      }\n    });\n  }\n  \n  grouped[groupKey]._count++;\n  \n  Object.entries(item).forEach(([key, value]) => {\n    if (key === groupByParam) return;\n    \n    if (value !== null && value !== undefined && value !== '' && grouped[groupKey]._values[key]) {\n      grouped[groupKey]._values[key].push(value);\n    }\n  });\n });\n\n const result = [];\n\n Object.entries(grouped).forEach(([groupKey, groupData]) => {\n  const aggregated = {\n    [groupByParam]: groupKey,\n    'Element Count': groupData._count\n  };\n  \n  Object.entries(groupData._values).forEach(([param, values]) => {\n    const rule = aggregationRules[param] || 'first';\n    \n    if (values.length === 0) {\n      aggregated[param] = null;\n      return;\n    }\n    \n    switch(rule) {\n      case 'sum':\n        const numericValues = values.map(v => {\n          const num = parseFloat(String(v).replace(',', '.'));\n          return isNaN(num) ? 0 : num;\n        });\n        aggregated[param] = numericValues.reduce((a, b) => a + b, 0);\n        // \u041e\u043a\u0440\u0443\u0433\u043b\u044f\u0435\u043c \u0434\u043e 2 \u0437\u043d\u0430\u043a\u043e\u0432 \u043f\u043e\u0441\u043b\u0435 \u0437\u0430\u043f\u044f\u0442\u043e\u0439 \u0435\u0441\u043b\u0438 \u043d\u0443\u0436\u043d\u043e\n        if (aggregated[param] % 1 !== 0) {\n          aggregated[param] = Math.round(aggregated[param] * 100) / 100;\n        }\n        break;\n        \n      case 'mean':\n      case 'average':\n        const avgValues = values.map(v => {\n          const num = parseFloat(String(v).replace(',', '.'));\n          return isNaN(num) ? null : num;\n        }).filter(v => v !== null);\n        \n        if (avgValues.length > 0) {\n          const avg = avgValues.reduce((a, b) => a + b, 0) / avgValues.length;\n          aggregated[param] = Math.round(avg * 100) / 100;\n        } else {\n          aggregated[param] = values[0];\n        }\n        break;\n        \n      case 'first':\n      default:\n        aggregated[param] = values[0];\n        break;\n    }\n  });\n  \n  result.push({ json: aggregated });\n });\n\n// \u0421\u043e\u0440\u0442\u0438\u0440\u0443\u0435\u043c \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\n result.sort((a, b) => {\n  const aVal = a.json[groupByParam];\n  const bVal = b.json[groupByParam];\n  if (aVal < bVal) return -1;\n  if (aVal > bVal) return 1;\n  return 0;\n });\n\n console.log(`\\nGrouping complete:`);\n console.log(`- Input items: ${cleanedData.length}`);\n console.log(`- Output groups: ${result.length}`);\n console.log(`- Parameters processed: ${Object.keys(aggregationRules).length}`);\n\n const rulesSummary = { sum: [], mean: [], first: [] };\n Object.entries(aggregationRules).forEach(([param, rule]) => {\n  if (rulesSummary[rule]) rulesSummary[rule].push(param);\n });\n\n console.log('\\nAggregation summary:');\n if (rulesSummary.sum.length > 0) {\n  console.log(`- SUM (${rulesSummary.sum.length}): ${rulesSummary.sum.slice(0, 5).join(', ')}${rulesSummary.sum.length > 5 ? '...' : ''}`);\n }\n if (rulesSummary.mean.length > 0) {\n  console.log(`- MEAN (${rulesSummary.mean.length}): ${rulesSummary.mean.slice(0, 5).join(', ')}${rulesSummary.mean.length > 5 ? '...' : ''}`);\n }\n if (rulesSummary.first.length > 0) {\n  console.log(`- FIRST (${rulesSummary.first.length}): ${rulesSummary.first.slice(0, 5).join(', ')}${rulesSummary.first.length > 5 ? '...' : ''}`);\n }\n\n return result;"
      },
      "typeVersion": 2
    },
    {
      "id": "ff967309-0ef4-4b49-abf1-b4c3ff6d48ac",
      "name": "AI Classify Categories1",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "position": [
        112,
        768
      ],
      "parameters": {
        "modelId": {
          "__rl": true,
          "mode": "list",
          "value": "chatgpt-4o-latest",
          "cachedResultName": "CHATGPT-4O-LATEST"
        },
        "options": {
          "maxTokens": 4000,
          "temperature": 0.1
        },
        "messages": {
          "values": [
            {
              "role": "system",
              "content": "You are an expert in Revit, BIM (Building Information Modeling) and construction classification. Your task is to classify category values as either building elements or non-building elements (drawings, annotations, etc.).\n\nBuilding elements include:\n- Structural elements (walls, floors, roofs, columns, beams, foundations, slabs)\n- MEP elements (pipes, ducts, equipment, fixtures, mechanical equipment)\n- Architectural elements (doors, windows, stairs, railings, curtain walls)\n- Site elements (parking, roads, landscaping)\n- Furniture and fixtures\n- Any physical construction element that has volume, area, or physical properties\n\nNon-building elements include:\n- Drawings and sheets\n- Annotations, dimensions, text notes\n- Views, sections, elevations, plans\n- Tags, symbols, legends, schedules\n- Grids, levels, reference planes\n- Revision clouds, callouts, detail items\n- Lines, filled regions, detail lines\n- Any 2D documentation or annotation element\n\nIMPORTANT: Analyze the actual category name, not just keywords. For example:\n- \"Detail Items\" = non-building (annotation)\n- \"Plumbing Fixtures\" = building element\n- \"Room Tags\" = non-building (annotation)\n- \"Structural Columns\" = building element\n\nReturn ONLY valid JSON in this exact format:\n{\n  \"classifications\": {\n    \"category_value_1\": true,\n    \"category_value_2\": false\n  },\n  \"building_categories\": [\"list\", \"of\", \"building\", \"categories\"],\n  \"drawing_categories\": [\"list\", \"of\", \"drawing\", \"categories\"]\n}"
            },
            {
              "content": "{{ $json.categoryField ? `Classify these ${$json.categoryValues.length} category values from field '${$json.categoryField}' as building elements (true) or drawings/annotations (false):\n\nCategory values:\n${JSON.stringify($json.categoryValues, null, 2)}\n\nCategory field type: ${$json.categoryFieldType}` : 'No category field found. Please classify based on volumetric data presence.'}}"
            }
          ]
        }
      },
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.3
    },
    {
      "id": "3e9cd8b6-3117-493b-97f7-c9d6a889e55d",
      "name": "Is Building Element1",
      "type": "n8n-nodes-base.if",
      "position": [
        640,
        768
      ],
      "parameters": {
        "conditions": {
          "boolean": [
            {
              "value1": "={{ $json.is_building_element }}",
              "value2": true
            }
          ]
        }
      },
      "typeVersion": 1
    },
    {
      "id": "1f3e2b55-6c9d-4466-801b-d8b31792a907",
      "name": "Check If All Batches Done",
      "type": "n8n-nodes-base.if",
      "position": [
        -416,
        1200
      ],
      "parameters": {
        "options": {},
        "conditions": {
          "options": {
            "leftValue": "",
            "caseSensitive": true,
            "typeValidation": "strict"
          },
          "combinator": "and",
          "conditions": [
            {
              "id": "9af8b6d4-3a3a-4c4a-8d4d-d5a8c2958f5f",
              "operator": {
                "type": "boolean",
                "operation": "true",
                "singleValue": true
              },
              "leftValue": "={{ $node['Process in Batches1'].context['noItemsLeft'] }}",
              "rightValue": true
            }
          ]
        }
      },
      "typeVersion": 2
    },
    {
      "id": "15da077e-8560-4bf9-b610-0199f66292f8",
      "name": "Collect All Results",
      "type": "n8n-nodes-base.code",
      "position": [
        -192,
        1552
      ],
      "parameters": {
        "jsCode": "// Get all accumulated data and prepare for reporting\nconst storedData = $getWorkflowStaticData('global');\nconst allProcessedItems = storedData.accumulatedData || [];\n\nconsole.log(`\\n=== BATCH PROCESSING COMPLETE ===`);\nconsole.log(`Total items processed: ${allProcessedItems.length}`);\n\n// Clear accumulated data for next run\nstoredData.accumulatedData = [];\n\n// Return all items for report generation\nreturn allProcessedItems;"
      },
      "typeVersion": 2
    },
    {
      "id": "7ae11a97-c8f9-4554-a1ae-0fc22c5ef6f9",
      "name": "Process in Batches1",
      "type": "n8n-nodes-base.splitInBatches",
      "position": [
        -256,
        1072
      ],
      "parameters": {
        "options": {},
        "batchSize": 1
      },
      "typeVersion": 1
    },
    {
      "id": "b2423ed4-6d42-4b31-a1c1-db9affbf720b",
      "name": "Clean Empty Values1",
      "type": "n8n-nodes-base.code",
      "position": [
        -80,
        1072
      ],
      "parameters": {
        "jsCode": "// Clean empty values from the item\nreturn $input.all().map(item => {\n  const cleanedJson = {};\n  Object.entries(item.json).forEach(([key, value]) => {\n    if (value !== null) {\n      if (typeof value === 'number') {\n        if (value !== 0 || key === 'Element Count') {\n          cleanedJson[key] = value;\n        }\n      } else if (value !== '') {\n        cleanedJson[key] = value;\n      }\n    }\n  });\n  return { json: cleanedJson };\n});"
      },
      "typeVersion": 2
    },
    {
      "id": "b9e77fc6-907b-47d2-8625-fdc0f6febb56",
      "name": "AI Agent Enhanced",
      "type": "@n8n/n8n-nodes-langchain.agent",
      "position": [
        272,
        1072
      ],
      "parameters": {
        "text": "={{ $json.userPrompt }}",
        "options": {
          "systemMessage": "={{ $json.systemPrompt }}"
        },
        "promptType": "define"
      },
      "typeVersion": 1.7
    },
    {
      "id": "2c3a57c2-66c2-414a-82e0-8f0df26cbfc8",
      "name": "Accumulate Results",
      "type": "n8n-nodes-base.code",
      "position": [
        848,
        1216
      ],
      "parameters": {
        "jsCode": "// Accumulate all processed items\nconst currentItem = $input.first();\nconst storedData = $getWorkflowStaticData('global');\n\n// Initialize accumulator if needed\nif (!storedData.accumulatedData) {\n  storedData.accumulatedData = [];\n}\n\n// Add current item to accumulated data\nstoredData.accumulatedData.push(currentItem);\n\nconsole.log(`Item processed. Total accumulated: ${storedData.accumulatedData.length} items`);\n\n// Return the current item to continue the flow\nreturn [currentItem];"
      },
      "typeVersion": 2
    },
    {
      "id": "fbf5d0a1-ed19-49e0-9a6e-27bf506bedec",
      "name": "Anthropic Chat Model1",
      "type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
      "position": [
        160,
        1264
      ],
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "claude-opus-4-20250514",
          "cachedResultName": "Claude Opus 4"
        },
        "options": {}
      },
      "credentials": {
        "anthropicApi": {
          "name": "<your credential>"
        }
      },
      "typeVersion": 1.3
    },
    {
      "id": "e7df68eb-4420-48b5-8465-b80c49f0122f",
      "name": "Calculate Project Totals4",
      "type": "n8n-nodes-base.code",
      "position": [
        -16,
        1552
      ],
      "parameters": {
        "jsCode": "// Aggregate all results and calculate project totals\nconst items = $input.all();\n\n// Initialize aggregators\nconst projectTotals = {\n  totalElements: 0,\n  totalMass: 0,\n  totalCO2: 0,\n  totalVolume: 0,\n  totalArea: 0,\n  byMaterial: {},\n  byCategory: {},\n  byImpact: {}\n};\n\n// Process each item\nitems.forEach(item => {\n  const data = item.json;\n  const elementCount = parseFloat(data['Element Count']) || 1;\n  const mass = parseFloat(data['Element Mass (tonnes)']) || 0;\n  const co2 = parseFloat(data['Total CO2 (tonnes CO2e)']) || 0;\n  const volume = parseFloat(data['Volume (m\u00b3)']) || 0;\n  const area = parseFloat(data['Area (m\u00b2)']) || 0;\n  const material = data['Material (EU Standard)'] || 'Unknown';\n  const category = data['Element Category'] || 'Unknown';\n  const impact = data['Impact Category'] || 'Unknown';\n  \n  // Update totals\n  projectTotals.totalElements += elementCount;\n  projectTotals.totalMass += mass;\n  projectTotals.totalCO2 += co2;\n  projectTotals.totalVolume += volume;\n  projectTotals.totalArea += area;\n  \n  // Aggregate by material\n  if (!projectTotals.byMaterial[material]) {\n    projectTotals.byMaterial[material] = {\n      elements: 0,\n      mass: 0,\n      co2: 0,\n      volume: 0,\n      types: new Set()\n    };\n  }\n  projectTotals.byMaterial[material].elements += elementCount;\n  projectTotals.byMaterial[material].mass += mass;\n  projectTotals.byMaterial[material].co2 += co2;\n  projectTotals.byMaterial[material].volume += volume;\n  projectTotals.byMaterial[material].types.add(data['Element Name']);\n  \n  // Aggregate by category\n  if (!projectTotals.byCategory[category]) {\n    projectTotals.byCategory[category] = {\n      elements: 0,\n      co2: 0\n    };\n  }\n  projectTotals.byCategory[category].elements += elementCount;\n  projectTotals.byCategory[category].co2 += co2;\n  \n  // Aggregate by impact\n  if (!projectTotals.byImpact[impact]) {\n    projectTotals.byImpact[impact] = {\n      elements: 0,\n      co2: 0\n    };\n  }\n  projectTotals.byImpact[impact].elements += elementCount;\n  projectTotals.byImpact[impact].co2 += co2;\n});\n\n// Add percentages and rankings to each item\nconst enrichedItems = items.map((item, index) => {\n  const data = item.json;\n  const co2 = parseFloat(data['Total CO2 (tonnes CO2e)']) || 0;\n  const mass = parseFloat(data['Element Mass (tonnes)']) || 0;\n  const elementCount = parseFloat(data['Element Count']) || 1;\n  \n  return {\n    json: {\n      ...data,\n      // Add project percentages\n      'CO2 % of Total': projectTotals.totalCO2 > 0 ? \n        ((co2 / projectTotals.totalCO2) * 100).toFixed(2) : '0.00',\n      'Mass % of Total': projectTotals.totalMass > 0 ? \n        ((mass / projectTotals.totalMass) * 100).toFixed(2) : '0.00',\n      'Elements % of Total': projectTotals.totalElements > 0 ? \n        ((elementCount / projectTotals.totalElements) * 100).toFixed(2) : '0.00',\n      // Add ranking\n      'CO2 Rank': index + 1,\n      // Project totals (same for all rows)\n      'Project Total Elements': projectTotals.totalElements,\n      'Project Total Mass (tonnes)': projectTotals.totalMass.toFixed(3),\n      'Project Total CO2 (tonnes)': projectTotals.totalCO2.toFixed(3),\n      'Project Total Volume (m\u00b3)': projectTotals.totalVolume.toFixed(2),\n      'Project Total Area (m\u00b2)': projectTotals.totalArea.toFixed(2)\n    }\n  };\n});\n\n// Sort by CO2 emissions (highest first)\nenrichedItems.sort((a, b) => \n  parseFloat(b.json['Total CO2 (tonnes CO2e)']) - parseFloat(a.json['Total CO2 (tonnes CO2e)'])\n);\n\n// Store aggregated data for summary\n$getWorkflowStaticData('global').projectTotals = projectTotals;\n\nreturn enrichedItems;"
      },
      "typeVersion": 2
    },
    {
      "id": "82b3f0eb-07d8-4a97-82a0-bbc38c7d040a",
      "name": "Enhance Excel Output",
      "type": "n8n-nodes-base.code",
      "position": [
        528,
        1712
      ],
      "parameters": {
        "jsCode": "// Enhanced Excel styling configuration\nconst excelBuffer = $input.first().binary.data;\nconst fileName = `CO2_Analysis_Professional_Report_${new Date().toISOString().slice(0,10)}.xlsx`;\n\n// Add metadata to the file\nconst metadata = {\n  title: 'Carbon Footprint Analysis Report',\n  author: 'DataDrivenConstruction.io',\n  company: 'Automated CO2 Analysis System',\n  created: new Date().toISOString(),\n  description: 'Comprehensive embodied carbon assessment with multi-standard material classification',\n  keywords: 'CO2, Carbon Footprint, Embodied Carbon, LCA, Building Materials'\n};\n\n// Return the enhanced Excel file\nreturn [{\n  json: {\n    fileName: fileName,\n    fileSize: excelBuffer.data.length,\n    sheets: 8,\n    metadata: metadata,\n    timestamp: new Date().toISOString()\n  },\n  binary: {\n    data: {\n      ...excelBuffer,\n      fileName: fileName,\n      fileExtension: 'xlsx',\n      mimeType: 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet'\n    }\n  }\n}];"
      },
      "typeVersion": 2
    },
    {
      "id": "4d379a3c-ab28-450b-a7a1-0eb28b706376",
      "name": "Prepare Excel Data",
      "type": "n8n-nodes-base.code",
      "position": [
        192,
        1712
      ],
      "parameters": {
        "jsCode": "// Prepare comprehensive data for multi-sheet Excel export\nconst items = $input.all();\nconst projectTotals = $getWorkflowStaticData('global').projectTotals;\n\n// Helper function to format numbers\nconst formatNumber = (num, decimals = 2) => {\n  return typeof num === 'number' ? parseFloat(num.toFixed(decimals)) : 0;\n};\n\n// Sheet 1: Executive Summary with project metrics\nconst executiveSummary = [\n  {\n    'Category': 'PROJECT OVERVIEW',\n    'Metric': 'Total Elements Analyzed',\n    'Value': projectTotals.totalElements,\n    'Unit': 'elements',\n    'Benchmark': 'N/A',\n    'Status': '\u2713',\n    'Notes': 'All building elements included in analysis'\n  },\n  {\n    'Category': 'PROJECT OVERVIEW',\n    'Metric': 'Element Groups',\n    'Value': items.length,\n    'Unit': 'groups',\n    'Benchmark': 'N/A',\n    'Status': '\u2713',\n    'Notes': 'Grouped by element type'\n  },\n  {\n    'Category': 'CARBON METRICS',\n    'Metric': 'Total Embodied Carbon',\n    'Value': formatNumber(projectTotals.totalCO2, 2),\n    'Unit': 'tonnes CO2e',\n    'Benchmark': 'Industry avg: ' + formatNumber(projectTotals.totalCO2 * 1.2, 0),\n    'Status': projectTotals.totalCO2 < projectTotals.totalCO2 * 1.2 ? '\u2713' : '\u26a0',\n    'Notes': 'A1-A3 lifecycle stages only'\n  },\n  {\n    'Category': 'CARBON METRICS',\n    'Metric': 'Average Carbon Intensity',\n    'Value': formatNumber(projectTotals.totalCO2 / projectTotals.totalMass, 3),\n    'Unit': 'kg CO2e/kg',\n    'Benchmark': '1.5-2.5',\n    'Status': (projectTotals.totalCO2 / projectTotals.totalMass) < 2.5 ? '\u2713' : '\u26a0',\n    'Notes': 'Average across all materials'\n  },\n  {\n    'Category': 'MATERIAL METRICS',\n    'Metric': 'Total Material Mass',\n    'Value': formatNumber(projectTotals.totalMass, 2),\n    'Unit': 'tonnes',\n    'Benchmark': 'N/A',\n    'Status': '\u2713',\n    'Notes': 'Combined mass of all materials'\n  },\n  {\n    'Category': 'MATERIAL METRICS',\n    'Metric': 'Unique Material Types',\n    'Value': Object.keys(projectTotals.byMaterial).length,\n    'Unit': 'materials',\n    'Benchmark': '10-20',\n    'Status': '\u2713',\n    'Notes': 'Distinct material classifications'\n  },\n  {\n    'Category': 'VOLUMETRIC DATA',\n    'Metric': 'Total Volume',\n    'Value': formatNumber(projectTotals.totalVolume, 2),\n    'Unit': 'm\u00b3',\n    'Benchmark': 'N/A',\n    'Status': '\u2713',\n    'Notes': 'Where volume data available'\n  },\n  {\n    'Category': 'VOLUMETRIC DATA',\n    'Metric': 'Total Area',\n    'Value': formatNumber(projectTotals.totalArea, 2),\n    'Unit': 'm\u00b2',\n    'Benchmark': 'N/A',\n    'Status': '\u2713',\n    'Notes': 'Where area data available'\n  }\n];\n\n// Sheet 2: Detailed Elements Analysis - ALL fields from processing\nconst detailedElements = items.map((item, index) => {\n  const data = item.json;\n  return {\n    // Ranking and identification\n    'CO2_Rank': index + 1,\n    'Element_Group': data['Element Name'] || data['Type Name'] || 'Unknown',\n    'Element_Category': data['Element Category'] || 'Unknown',\n    'Element_Type': data['Element Type'] || 'Unknown',\n    'Element_Function': data['Element Function'] || 'Unknown',\n    'Element_Count': parseInt(data['Element Count']) || 0,\n    \n    // Material classification (all 3 standards)\n    'Material_EU_Standard': data['Material (EU Standard)'] || 'Unknown',\n    'Material_DE_Standard': data['Material (DE Standard)'] || 'Unknown',\n    'Material_US_Standard': data['Material (US Standard)'] || 'Unknown',\n    'Primary_Material': data['Primary Material'] || 'Unknown',\n    'Secondary_Materials': data['Secondary Materials'] || 'None',\n    \n    // Quantities and dimensions\n    'Quantity_Value': formatNumber(data['Quantity'] || 0, 3),\n    'Quantity_Unit': data['Quantity Unit'] || 'piece',\n    'Volume_m3': formatNumber(data['Volume (m\u00b3)'] || 0, 3),\n    'Area_m2': formatNumber(data['Area (m\u00b2)'] || 0, 3),\n    'Length_mm': formatNumber(data['Length (mm)'] || 0, 0),\n    'Width_mm': formatNumber(data['Width (mm)'] || 0, 0),\n    'Height_mm': formatNumber(data['Height (mm)'] || 0, 0),\n    'Thickness_mm': formatNumber(data['Thickness (mm)'] || 0, 0),\n    \n    // Mass and density\n    'Material_Density_kg_m3': formatNumber(data['Material Density (kg/m\u00b3)'] || 0, 0),\n    'Element_Mass_kg': formatNumber(data['Element Mass (kg)'] || 0, 2),\n    'Element_Mass_tonnes': formatNumber(data['Element Mass (tonnes)'] || 0, 3),\n    \n    // CO2 emissions data\n    'CO2_Factor_kg_CO2_per_kg': formatNumber(data['CO2 Factor (kg CO2e/kg)'] || 0, 3),\n    'Total_CO2_kg': formatNumber(data['Total CO2 (kg CO2e)'] || 0, 2),\n    'Total_CO2_tonnes': formatNumber(data['Total CO2 (tonnes CO2e)'] || 0, 3),\n    'CO2_per_Element_kg': formatNumber(data['CO2 per Element (kg CO2e)'] || 0, 2),\n    'CO2_Intensity': formatNumber(data['CO2 Intensity'] || 0, 3),\n    'CO2_Percent_of_Total': formatNumber(data['CO2 % of Total'] || 0, 2),\n    \n    // Impact and quality metrics\n    'Impact_Category': data['Impact Category'] || 'Unknown',\n    'Lifecycle_Stage': data['Lifecycle Stage'] || 'A1-A3',\n    'Data_Source': data['Data Source'] || 'Industry average',\n    \n    // Confidence scores\n    'Overall_Confidence_%': parseInt(data['Overall Confidence (%)']) || 0,\n    'Material_Confidence_%': parseInt(data['Material Confidence (%)']) || 0,\n    'Quantity_Confidence_%': parseInt(data['Quantity Confidence (%)']) || 0,\n    'CO2_Confidence_%': parseInt(data['CO2 Confidence (%)']) || 0,\n    'Data_Quality': data['Data Quality'] || 'unknown',\n    \n    // Analysis metadata\n    'Calculation_Method': data['Calculation Method'] || 'Not specified',\n    'Assumptions': data['Assumptions'] || 'None',\n    'Warnings': data['Warnings'] || 'None',\n    'Analysis_Notes': data['Analysis Notes'] || '',\n    'Processing_Timestamp': data['Processing Timestamp'] || new Date().toISOString(),\n    'Analysis_Status': data['Analysis Status'] || 'Unknown'\n  };\n});\n\n// Sheet 3: Material Summary with detailed breakdown\nconst materialSummary = Object.entries(projectTotals.byMaterial)\n  .sort((a, b) => b[1].co2 - a[1].co2)\n  .map(([material, data], index) => {\n    const co2Percent = (data.co2 / projectTotals.totalCO2) * 100;\n    const massPercent = (data.mass / projectTotals.totalMass) * 100;\n    const avgCO2Factor = data.mass > 0 ? data.co2 / data.mass : 0;\n    \n    return {\n      'Rank': index + 1,\n      'Material_Type': material,\n      'Element_Count': data.elements,\n      'Unique_Types': data.types ? data.types.size : 0,\n      'Mass_tonnes': formatNumber(data.mass, 2),\n      'Mass_%': formatNumber(massPercent, 1),\n      'Volume_m3': formatNumber(data.volume, 2),\n      'CO2_tonnes': formatNumber(data.co2, 2),\n      'CO2_%': formatNumber(co2Percent, 1),\n      'Avg_CO2_Factor': formatNumber(avgCO2Factor, 3),\n      'CO2_per_Element_kg': formatNumber(data.co2 / data.elements * 1000, 1),\n      'Impact_Level': co2Percent >= 20 ? 'CRITICAL' : co2Percent >= 10 ? 'HIGH' : co2Percent >= 5 ? 'MEDIUM' : 'LOW',\n      'Reduction_Potential_20%': formatNumber(data.co2 * 0.2, 2),\n      'Benchmark_Factor': formatNumber(avgCO2Factor * 0.8, 3)\n    };\n  });\n\n// Sheet 4: Category Analysis\nconst categoryAnalysis = Object.entries(projectTotals.byCategory)\n  .sort((a, b) => b[1].co2 - a[1].co2)\n  .map(([category, data], index) => {\n    const co2Percent = (data.co2 / projectTotals.totalCO2) * 100;\n    const elementsPercent = (data.elements / projectTotals.totalElements) * 100;\n    \n    return {\n      'Rank': index + 1,\n      'Category': category,\n      'Element_Count': data.elements,\n      'Elements_%': formatNumber(elementsPercent, 1),\n      'CO2_tonnes': formatNumber(data.co2, 2),\n      'CO2_%': formatNumber(co2Percent, 1),\n      'Avg_CO2_per_Element': formatNumber(data.co2 / data.elements * 1000, 1),\n      'CO2_Intensity_Ratio': formatNumber((data.co2 / data.elements) / (projectTotals.totalCO2 / projectTotals.totalElements), 2),\n      'Priority': co2Percent >= 15 ? 'HIGH' : co2Percent >= 5 ? 'MEDIUM' : 'LOW'\n    };\n  });\n\n// Sheet 5: Impact Analysis by Category\nconst impactAnalysis = Object.entries(projectTotals.byImpact || {})\n  .map(([impact, data]) => ({\n    'Impact_Category': impact,\n    'Element_Count': data.elements,\n    'CO2_tonnes': formatNumber(data.co2, 2),\n    'CO2_%': formatNumber((data.co2 / projectTotals.totalCO2) * 100, 1),\n    'Avg_CO2_per_Element': formatNumber(data.co2 / data.elements * 1000, 1)\n  }));\n\n// Sheet 6: Top 20 Hotspots with action items\nconst top20Hotspots = items\n  .slice(0, 20)\n  .map((item, index) => {\n    const data = item.json;\n    const co2Tonnes = parseFloat(data['Total CO2 (tonnes CO2e)']) || 0;\n    const co2Percent = parseFloat(data['CO2 % of Total']) || 0;\n    const elementCount = parseInt(data['Element Count']) || 1;\n    \n    // Generate specific recommendations based on material and impact\n    let recommendation = '';\n    if (co2Percent >= 10) {\n      recommendation = 'CRITICAL: Prioritize immediate material substitution or design optimization';\n    } else if (co2Percent >= 5) {\n      recommendation = 'HIGH: Evaluate low-carbon alternatives and quantity reduction opportunities';\n    } else if (co2Percent >= 2) {\n      recommendation = 'MEDIUM: Consider optimization during value engineering phase';\n    } else {\n      recommendation = 'LOW: Monitor and optimize if convenient';\n    }\n    \n    return {\n      'Priority_Rank': index + 1,\n      'Element_Group': data['Element Name'] || 'Unknown',\n      'Category': data['Element Category'] || 'Unknown',\n      'Material': data['Material (EU Standard)'] || 'Unknown',\n      'Element_Count': elementCount,\n      'Mass_tonnes': formatNumber(data['Element Mass (tonnes)'] || 0, 2),\n      'CO2_tonnes': formatNumber(co2Tonnes, 3),\n      'CO2_%': formatNumber(co2Percent, 2),\n      'CO2_per_Element': formatNumber(co2Tonnes / elementCount * 1000, 1),\n      'Impact_Level': co2Percent >= 10 ? 'CRITICAL' : co2Percent >= 5 ? 'HIGH' : co2Percent >= 2 ? 'MEDIUM' : 'LOW',\n      'Confidence_%': parseInt(data['Overall Confidence (%)']) || 0,\n      'Reduction_Target_20%': formatNumber(co2Tonnes * 0.2, 2),\n      'Recommendation': recommendation\n    };\n  });\n\n// Sheet 7: Data Quality Report\nconst dataQuality = items.map((item, index) => {\n  const data = item.json;\n  return {\n    'Element_Rank': index + 1,\n    'Element_Group': data['Element Name'] || 'Unknown',\n    'Overall_Confidence_%': parseInt(data['Overall Confidence (%)']) || 0,\n    'Material_Confidence_%': parseInt(data['Material Confidence (%)']) || 0,\n    'Quantity_Confidence_%': parseInt(data['Quantity Confidence (%)']) || 0,\n    'CO2_Confidence_%': parseInt(data['CO2 Confidence (%)']) || 0,\n    'Data_Quality': data['Data Quality'] || 'unknown',\n    'Data_Source': data['Data Source'] || 'Unknown',\n    'Assumptions': data['Assumptions'] || 'None',\n    'Warnings': data['Warnings'] || 'None',\n    'Analysis_Status': data['Analysis Status'] || 'Unknown'\n  };\n}).filter(item => \n  item['Overall_Confidence_%'] < 90 || \n  item['Data_Quality'] !== 'high' || \n  item['Warnings'] !== 'None'\n);\n\n// Sheet 8: Recommendations Summary\nconst recommendations = [\n  {\n    'Priority': 1,\n    'Category': 'IMMEDIATE ACTIONS',\n    'Recommendation': `Focus on ${Object.entries(projectTotals.byMaterial).sort((a,b) => b[1].co2 - a[1].co2)[0][0]} optimization`,\n    'Potential_Savings': formatNumber(Object.entries(projectTotals.byMaterial).sort((a,b) => b[1].co2 - a[1].co2)[0][1].co2 * 0.2, 1) + ' tonnes CO2e',\n    'Implementation': 'Material substitution or design optimization',\n    'Timeline': '0-3 months'\n  },\n  {\n    'Priority': 2,\n    'Category': 'IMMEDIATE ACTIONS',\n    'Recommendation': `Review top ${items.filter(item => parseFloat(item.json['CO2 % of Total']) >= 5).length} high-impact element groups`,\n    'Potential_Savings': formatNumber(projectTotals.totalCO2 * 0.15, 1) + ' tonnes CO2e',\n    'Implementation': 'Design review and value engineering',\n    'Timeline': '0-3 months'\n  },\n  {\n    'Priority': 3,\n    'Category': 'SHORT TERM',\n    'Recommendation': 'Implement low-carbon concrete mixes where applicable',\n    'Potential_Savings': '10-15% reduction possible',\n    'Implementation': 'Specification updates',\n    'Timeline': '3-6 months'\n  },\n  {\n    'Priority': 4,\n    'Category': 'SHORT TERM',\n    'Recommendation': 'Increase recycled content in steel elements',\n    'Potential_Savings': '20-30% reduction possible',\n    'Implementation': 'Supplier engagement',\n    'Timeline': '3-6 months'\n  },\n  {\n    'Priority': 5,\n    'Category': 'MEDIUM TERM',\n    'Recommendation': 'Explore timber alternatives for suitable applications',\n    'Potential_Savings': 'Carbon negative potential',\n    'Implementation': 'Structural analysis required',\n    'Timeline': '6-12 months'\n  }\n];\n\n// Create worksheet structure with proper sheet names\nconst worksheets = [\n  { name: 'Executive Summary', data: executiveSummary },\n  { name: 'All Elements', data: detailedElements },\n  { name: 'Material Summary', data: materialSummary },\n  { name: 'Category Analysis', data: categoryAnalysis },\n  { name: 'Impact Analysis', data: impactAnalysis },\n  { name: 'Top 20 Hotspots', data: top20Hotspots },\n  { name: 'Data Quality', data: dataQuality },\n  { name: 'Recommendations', data: recommendations }\n];\n\n// Flatten all data with sheet markers\nconst allData = [];\nworksheets.forEach(sheet => {\n  sheet.data.forEach(row => {\n    allData.push({\n      json: {\n        ...row,\n        _sheetName: sheet.name\n      }\n    });\n  });\n});\n\nreturn allData;"
      },
      "typeVersion": 2
    },
    {
      "id": "a4d43d4c-ef26-4d96-b8df-88cdff94a9b3",
      "name": "Create Excel File",
      "type": "n8n-nodes-base.spreadsheetFile",
      "position": [
        352,
        1712
      ],
      "parameters": {
        "options": {
          "fileName": "=CO2_Analysis_Report_{{ $now.format('yyyy-MM-dd') }}",
          "headerRow": true,
          "sheetName": "={{ $json._sheetName }}"
        },
        "operation": "toFile",
        "fileFormat": "xlsx"
      },
      "typeVersion": 2
    },
    {
      "id": "36bb02f7-baba-48c7-aad7-4fe2b057e6be",
      "name": "Prepare Enhanced Prompts",
      "type": "n8n-nodes-base.code",
      "position": [
        96,
        1072
      ],
      "parameters": {
        "jsCode": "// Enhanced prompts for comprehensive CO2 analysis\nconst inputData = $input.first().json;\nconst originalGroupedData = { ...inputData };\n\nconst systemPrompt = `You are an expert in construction materials, carbon footprint analysis, and building element classification. Analyze the provided building element data and return a comprehensive CO2 emissions assessment.\n\nIMPORTANT NOTE ON DATA:\n- All quantitative fields like Volume, Area, Length, etc. in the input data represent ALREADY AGGREGATED TOTALS for the entire group of elements.\n- 'Element Count' indicates the number of individual elements in this group.\n- DO NOT multiply volumes/areas by Element Count - they are already total sums.\n- Use the provided totals directly for mass and CO2 calculations.\n- If no volumetric data is available, estimate based on typical values, but prioritize provided data.\n\n## Your Analysis Must Include:\n\n### 1. Element Identification\n- Element type and category\n- Primary material composition\n- Secondary materials if applicable\n- Functional classification\n\n### 2. Material Classification (All 3 Standards)\n- **European (EN 15978/15804)**: YOUR_AWS_SECRET_KEY_HERE.\n- **German (\u00d6KOBAUDAT)**: Mineralische/Metalle/Holz/D\u00e4mmstoffe/etc.\n- **US (MasterFormat)**: Division classifications\n\n### 3. Quantity Analysis\n- Primary quantity with unit (m\u00b3, m\u00b2, m, kg, pieces) - use aggregated total from input\n- Calculation method used (e.g., 'Direct from provided total volume')\n- Confidence level (0-100%)\n- Data quality assessment\n\n### 4. CO2 Emissions Calculation\n- Material density (kg/m\u00b3)\n- Mass calculation: Use total volume/area * density (do not multiply by count)\n- CO2 emission factor (kg CO2e/kg)\n- Total CO2 emissions (for the entire group)\n- Emission intensity metrics\n\n### 5. Data Quality & Confidence\n- Overall confidence score\n- Data completeness assessment\n- Key assumptions made\n- Warnings or limitations\n\n## Important Guidelines:\n1. Use industry-standard emission factors\n2. Apply conservative estimates when uncertain\n3. Consider full lifecycle emissions (A1-A3 minimum)\n4. Account for regional variations where relevant\n5. Include embodied carbon only (not operational)\n\n## Output Format\nReturn ONLY valid JSON with this exact structure:\n{\n  \"element_identification\": {\n    \"name\": \"string\",\n    \"category\": \"string\",\n    \"type\": \"string\",\n    \"function\": \"string\"\n  },\n  \"material_classification\": {\n    \"european\": \"string\",\n    \"german\": \"string\",\n    \"us\": \"string\",\n    \"primary_material\": \"string\",\n    \"secondary_materials\": [\"string\"]\n  },\n  \"quantities\": {\n    \"value\": number,\n    \"unit\": \"string\",\n    \"calculation_method\": \"string\",\n    \"raw_dimensions\": {\n      \"length\": number,\n      \"width\": number,\n      \"height\": number,\n      \"thickness\": number,\n      \"area\": number,\n      \"volume\": number\n    }\n  },\n  \"co2_analysis\": {\n    \"density_kg_m3\": number,\n    \"mass_kg\": number,\n    \"co2_factor_kg_co2_per_kg\": number,\n    \"total_co2_kg\": number,\n    \"co2_intensity_kg_per_unit\": number,\n    \"lifecycle_stage\": \"A1-A3\",\n    \"data_source\": \"string\"\n  },\n  \"confidence\": {\n    \"overall_score\": number,\n    \"material_confidence\": number,\n    \"quantity_confidence\": number,\n    \"co2_confidence\": number,\n    \"data_quality\": \"high/medium/low\"\n  },\n  \"metadata\": {\n    \"assumptions\": [\"string\"],\n    \"warnings\": [\"string\"],\n    \"notes\": \"string\"\n  }\n}`;\n\nconst userPrompt = `Analyze this building element group for CO2 emissions. Remember: Volumes and areas are already total for the group, not per element.\n\n${JSON.stringify(inputData, null, 2)}\n\nProvide comprehensive CO2 analysis following the specified format. Focus on accuracy and use conservative estimates where data is uncertain.`;\n\nreturn [{\n  json: {\n    ...originalGroupedData,\n    systemPrompt,\n    userPrompt,\n    _originalGroupedData: originalGroupedData\n  }\n}];"
      },
      "typeVersion": 2
    },
    {
      "id": "32640ecd-b337-4745-bbe2-072e60163dcf",
      "name": "Parse Enhanced Response",
      "type": "n8n-nodes-base.code",
      "position": [
        624,
        1072
      ],
      "parameters": {
        "jsCode": "// Parse and enrich AI response with all necessary data\nconst aiResponse = $input.first().json.output || $input.first().json.response || $input.first().json.text || $input.first().json;\nconst originalData = $node[\"Prepare Enhanced Prompts\"].json._originalGroupedData || $node[\"Prepare Enhanced Prompts\"].json;\n\ntry {\n  // Extract JSON from response\n  let jsonStr = aiResponse;\n  if (typeof jsonStr === 'string') {\n    const jsonMatch = jsonStr.match(/```json\\n?([\\s\\S]*?)\\n?```/) || jsonStr.match(/\\{[\\s\\S]*\\}/);\n    if (jsonMatch) {\n      jsonStr = jsonMatch[1] || jsonMatch[0];\n    }\n  }\n  \n  const analysis = typeof jsonStr === 'string' ? JSON.parse(jsonStr) : jsonStr;\n  \n  // Calculate additional metrics - use total CO2 for group\n  const co2_kg = analysis.co2_analysis?.total_co2_kg || 0;\n  const co2_tonnes = co2_kg / 1000;\n  const elementCount = parseFloat(originalData['Element Count']) || 1;\n  const co2_per_element = co2_kg / elementCount;\n  \n  // Determine impact category\n  let impactCategory = 'Unknown';\n  const co2Factor = analysis.co2_analysis?.co2_factor_kg_co2_per_kg || 0;\n  if (co2Factor < 0) {\n    impactCategory = 'Carbon Negative (Storage)';\n  } else if (co2Factor <= 0.5) {\n    impactCategory = 'Very Low Impact';\n  } else if (co2Factor <= 1.0) {\n    impactCategory = 'Low Impact';\n  } else if (co2Factor <= 2.0) {\n    impactCategory = 'Medium Impact';\n  } else if (co2Factor <= 5.0) {\n    impactCategory = 'High Impact';\n  } else {\n    impactCategory = 'Very High Impact';\n  }\n  \n  // Create comprehensive output record\n  return [{\n    json: {\n      // Original data\n      ...originalData,\n      \n      // Element identification\n      'Element Name': analysis.element_identification?.name || originalData['Type Name'] || 'Unknown',\n      'Element Category': analysis.element_identification?.category || originalData['Category'] || 'Unknown',\n      'Element Type': analysis.element_identification?.type || 'Unknown',\n      'Element Function': analysis.element_identification?.function || 'Unknown',\n      \n      // Material classification\n      'Material (EU Standard)': analysis.material_classification?.european || 'Unknown',\n      'Material (DE Standard)': analysis.material_classification?.german || 'Unknown',\n      'Material (US Standard)': analysis.material_classification?.us || 'Unknown',\n      'Primary Material': analysis.material_classification?.primary_material || 'Unknown',\n      'Secondary Materials': (analysis.material_classification?.secondary_materials || []).join(', ') || 'None',\n      \n      // Quantities\n      'Quantity': analysis.quantities?.value || 0,\n      'Quantity Unit': analysis.quantities?.unit || 'piece',\n      'Calculation Method': analysis.quantities?.calculation_method || 'Not specified',\n      \n      // Dimensions (from raw data) - these are totals\n      'Length (mm)': analysis.quantities?.raw_dimensions?.length || originalData['Length'] || 0,\n      'Width (mm)': analysis.quantities?.raw_dimensions?.width || originalData['Width'] || 0,\n      'Height (mm)': analysis.quantities?.raw_dimensions?.height || originalData['Height'] || 0,\n      'Thickness (mm)': analysis.quantities?.raw_dimensions?.thickness || originalData['Thickness'] || 0,\n      'Area (m\u00b2)': analysis.quantities?.raw_dimensions?.area || originalData['Area'] || 0,\n      'Volume (m\u00b3)': analysis.quantities?.raw_dimensions?.volume || originalData['Volume'] || 0,\n      \n      // CO2 Analysis\n      'Material Density (kg/m\u00b3)': analysis.co2_analysis?.density_kg_m3 || 0,\n      'Element Mass (kg)': analysis.co2_analysis?.mass_kg || 0,\n      'Element Mass (tonnes)': (analysis.co2_analysis?.mass_kg || 0) / 1000,\n      'CO2 Factor (kg CO2e/kg)': analysis.co2_analysis?.co2_factor_kg_co2_per_kg || 0,\n      'Total CO2 (kg CO2e)': co2_kg,\n      'Total CO2 (tonnes CO2e)': co2_tonnes,\n      'CO2 per Element (kg CO2e)': co2_per_element,\n      'CO2 Intensity': analysis.co2_analysis?.co2_intensity_kg_per_unit || 0,\n      'Lifecycle Stage': analysis.co2_analysis?.lifecycle_stage || 'A1-A3',\n      'Data Source': analysis.co2_analysis?.data_source || 'Industry average',\n      'Impact Category': impactCategory,\n      \n      // Confidence scores\n      'Overall Confidence (%)': analysis.confidence?.overall_score || 0,\n      'Material Confidence (%)': analysis.confidence?.material_confidence || 0,\n      'Quantity Confidence (%)': analysis.confidence?.quantity_confidence || 0,\n      'CO2 Confidence (%)': analysis.confidence?.co2_confidence || 0,\n      'Data Quality': analysis.confidence?.data_quality || 'unknown',\n      \n      // Metadata\n      'Assumptions': (analysis.metadata?.assumptions || []).join('; ') || 'None',\n      'Warnings': (analysis.metadata?.warnings || []).join('; ') || 'None',\n      'Analysis Notes': analysis.metadata?.notes || '',\n      'Processing Timestamp': new Date().toISOString(),\n      'Analysis Status': 'Complete'\n    }\n  }];\n  \n} catch (error) {\n  // Return error record with original data preserved\n  return [{\n    json: {\n      ...originalData,\n      'Analysis Status': 'Failed',\n      'Error': error.message,\n      'Processing Timestamp': new Date().toISOString()\n    }\n  }];\n}"
      },
      "typeVersion": 2
    },
    {
      "id": "62b7db30-b6f2-4608-bcd8-f866759bcf56",
      "name": "Read Excel File",
      "type": "n8n-nodes-base.readBinaryFile",
      "position": [
        -288,
        496
      ],
      "parameters": {
        "filePath": "={{ $json.path_to_file }}"
      },
      "typeVersion": 1
    },
    {
      "id": "40d06cd0-e57e-4d9d-8557-a06f64126d43",
      "name": "Parse Excel",
      "type": "n8n-nodes-base.spreadsheetFile",
      "position": [
        -96,
        496
      ],
      "parameters": {
        "options": {
          "headerRow": true,
          "sheetName": "={{ $node['Set Parameters'].json.sheet_name }}",
          "includeEmptyCells": false
        },
        "fileFormat": "xlsx"
      },
      "typeVersion": 2
    },
    {
      "id": "5f385292-ebab-4c4f-92b9-15c7ee875e02",
      "name": "Setup - Define file paths",
      "type": "n8n-nodes-base.set",
      "position": [
        -272,
        -64
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "9cbd4ec9-df24-41e8-b47a-720a4cdb733b",
              "name": "path_to_converter",
              "type": "string",
              "value": "C:\\Users\\Artem Boiko\\Desktop\\n8n pipelines\\DDC_Converter_Revit\\datadrivenlibs\\RvtExporter.exe"
            },
            {
              "id": "aa834467-80fb-476a-bac1-6728478834f0",
              "name": "project_file",
              "type": "string",
              "value": "C:\\Users\\Artem Boiko\\Documents\\GitHub\\cad2data-Revit-IFC-DWG-DGN-pipeline-with-conversion-validation-qto\\Sample_Projects\\2023 racbasicsampleproject.rvt"
            },
            {
              "id": "4e4f5e6f-7a8b-4c5d-9e0f-1a2b3c4d5e6f",
              "name": "group_by",
              "type": "string",
              "value": "Type Name"
            },
            {
              "id": "5f6a7b8c-9d0e-4f1a-2b3c-4d5e6f7a8b9c",
              "name": "country",
              "type": "string",
              "value": "Germany"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "c8ae79e4-3a12-4b9e-bcc2-129adf5ecf25",
      "name": "Create - Excel filename",
      "type": "n8n-nodes-base.set",
      "position": [
        -48,
        -64
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "xlsx-filename-id",
              "name": "xlsx_filename",
              "type": "string",
              "value": "={{ $json[\"project_file\"].slice(0, -4) + \"_rvt.xlsx\" }}"
            },
            {
              "id": "path-to-converter-pass",
              "name": "path_to_converter",
              "type": "string",
              "value": "={{ $json[\"path_to_converter\"] }}"
            },
            {
              "id": "project-file-pass",
              "name": "project_file",
              "type": "string",
              "value": "={{ $json[\"project_file\"] }}"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "60bcb5b5-2f61-4314-94a2-7d0bf53427fe",
      "name": "Check - Does Excel file exist?",
      "type": "n8n-nodes-base.readBinaryFile",
      "position": [
        144,
        -64
      ],
      "parameters": {
        "filePath": "={{ $json[\"xlsx_filename\"] }}"
      },
      "typeVersion": 1,
      "continueOnFail": true,
      "alwaysOutputData": true
    },
    {
      "id": "4a82e1e2-4c87-4132-82f0-bfcacf07ca38",
      "name": "If - File exists?",
      "type": "n8n-nodes-base.if"

Credentials you'll need

Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.

• anthropicApi
• openAiApi
• xAiApi

About this workflow

Estimate embodied carbon (CO2e) for grouped BIM/CAD elements. The workflow accepts an existing XLSX (grouped element data) or, if missing, can trigger a local RvtExporter.exe to generate one. It detects category fields, filters out non-building elements, infers aggregation rules…

Source: https://n8n.io/workflows/7653/ — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.