bim-cost-estimation-cwicr

name	bim-cost-estimation-cwicr
description	Automated cost estimation from BIM models using DDC CWICR database with 55,719 work items. AI classification + vector search for accurate pricing.
homepage	https://datadrivenconstruction.io
metadata	{"openclaw":{"emoji":"🏗️","os":["darwin","linux","win32"],"homepage":"https://datadrivenconstruction.io","requires":{"bins":["python3"],"env":["OPENAI_API_KEY","QDRANT_URL"]},"primaryEnv":"OPENAI_API_KEY"}}

BIM Cost Estimation with DDC CWICR

Generate accurate cost estimates from BIM models using AI classification and the DDC CWICR construction cost database.

Business Case

Problem: Traditional cost estimation:

• Manual and time-consuming (weeks for detailed estimate)
• Subjective and inconsistent between estimators
• Requires specialized knowledge
• Difficult to update with design changes

Solution: Automated BIM-to-cost pipeline:

• Extract quantities directly from model
• AI classifies elements to work items
• Vector search finds matching prices in CWICR
• Complete estimate in hours, not weeks

ROI: 80% reduction in estimation time, consistent methodology

System Architecture

┌──────────────────────────────────────────────────────────────────────────┐
│                  BIM TO COST ESTIMATION PIPELINE                          │
├──────────────────────────────────────────────────────────────────────────┤
│                                                                           │
│   ┌─────────┐     ┌─────────┐     ┌─────────┐     ┌─────────────────┐   │
│   │ BIM     │     │ DDC     │     │ AI      │     │ DDC CWICR       │   │
│   │ Model   │────►│Converter│────►│ LLM     │────►│ Vector Search   │   │
│   │.rvt/.ifc│     │         │     │         │     │ (Qdrant)        │   │
│   └─────────┘     └─────────┘     └─────────┘     └─────────────────┘   │
│                        │              │                    │             │
│                        ▼              ▼                    ▼             │
│                   ┌─────────┐    ┌─────────┐         ┌──────────┐       │
│                   │ .xlsx   │    │ Work    │         │ Matched  │       │
│                   │ QTO     │    │ Items   │         │ Rates    │       │
│                   └─────────┘    └─────────┘         └──────────┘       │
│                        │              │                    │             │
│                        └──────────────┼────────────────────┘             │
│                                       ▼                                  │
│                              ┌─────────────────┐                        │
│                              │ COST ESTIMATE   │                        │
│                              │                 │                        │
│                              │ • By element    │                        │
│                              │ • By trade      │                        │
│                              │ • By phase      │                        │
│                              │ • Resources     │                        │
│                              └─────────────────┘                        │
│                                                                           │
└──────────────────────────────────────────────────────────────────────────┘

DDC CWICR Database

Database Overview:
  work_items: 55,719
  resources: 27,672
  languages: 9 (AR, DE, EN, ES, FR, HI, PT, RU, ZH)
  fields_per_item: 85
  embedding_model: text-embedding-3-large (3072d)
  vector_db: Qdrant

Collections:
  - ddc_cwicr_ar  # Arabic (Dubai prices)
  - ddc_cwicr_de  # German (Berlin prices)
  - ddc_cwicr_en  # English (Toronto prices)
  - ddc_cwicr_es  # Spanish (Barcelona prices)
  - ddc_cwicr_fr  # French (Paris prices)
  - ddc_cwicr_hi  # Hindi (Mumbai prices)
  - ddc_cwicr_pt  # Portuguese (São Paulo prices)
  - ddc_cwicr_ru  # Russian (St. Petersburg prices)
  - ddc_cwicr_zh  # Chinese (Shanghai prices)

Pipeline Stages

Stage	Name	Description
0	Collect BIM Data	Extract elements from Revit/IFC
1	Project Detection	AI identifies project type
2	Phase Generation	AI creates construction phases
3	Element Assignment	AI maps types to phases
4	Work Decomposition	AI breaks types into work items
5	Vector Search	Find matching rates in CWICR
6	Unit Mapping	Convert BIM units to rate units
7	Cost Calculation	Qty × Unit Price
7.5	Validation	CTO review for completeness
8	Aggregation	Sum by phases and categories
9	Report Generation	HTML and Excel outputs

Python Implementation

import pandas as pd
import numpy as np
from qdrant_client import QdrantClient
from qdrant_client.models import Filter, FieldCondition, MatchValue
from openai import OpenAI
from typing import List, Dict, Optional
from dataclasses import dataclass
import json

@dataclass
class WorkItem:
    """Matched work item from CWICR"""
    cwicr_code: str
    description: str
    unit: str
    unit_price: float
    labor_cost: float
    material_cost: float
    equipment_cost: float
    productivity: float  # units per hour
    currency: str
    confidence: float

@dataclass
class CostLineItem:
    """Single line item in estimate"""
    bim_type: str
    work_item: WorkItem
    quantity: float
    quantity_unit: str
    total_cost: float
    labor_cost: float
    material_cost: float
    equipment_cost: float
    phase: str
    trade: str


class :
    

     ():
        .qdrant = QdrantClient(url=qdrant_url, api_key=qdrant_api_key)
        .openai = OpenAI(api_key=openai_api_key)
        .language = language
        .collection = 

     () -> []:
        
        response = .openai.embeddings.create(
            model=,
            =text,
            dimensions=
        )
         response.data[].embedding

     () -> [WorkItem]:
        

        
        query_vector = .get_embedding(query)

        
        query_filter = 
         category_filter:
            query_filter = Filter(
                must=[
                    FieldCondition(
                        key=,
                        =MatchValue(value=category_filter)
                    )
                ]
            )

        
        results = .qdrant.search(
            collection_name=.collection,
            query_vector=query_vector,
            query_filter=query_filter,
            limit=limit
        )

        
        work_items = []
         r  results:
            payload = r.payload
            work_items.append(WorkItem(
                cwicr_code=payload.get(, ),
                description=payload.get(, ),
                unit=payload.get(, ),
                unit_price=(payload.get(, )),
                labor_cost=(payload.get(, )),
                material_cost=(payload.get(, )),
                equipment_cost=(payload.get(, )),
                productivity=(payload.get(, )),
                currency=payload.get(, ),
                confidence=r.score
            ))

         work_items

     () -> []:
        

        prompt = 

        response = .openai.chat.completions.create(
            model=,
            messages=[{: , : prompt}],
            response_format={: }
        )

        :
            result = json.loads(response.choices[].message.content)
             result.get(, [bim_type])
        :
             [bim_type]

     () -> [CostLineItem]:
        

        
        work_descriptions = .decompose_bim_type(bim_type, category)

        line_items = []

         work_desc  work_descriptions:
            
            matches = .search_cwicr(work_desc, limit=)

              matches:
                

            best_match = matches[]

            
            adjusted_qty = ._convert_units(
                quantity, quantity_unit, best_match.unit
            )

            
            total = adjusted_qty * best_match.unit_price
            labor = adjusted_qty * best_match.labor_cost
            material = adjusted_qty * best_match.material_cost
            equipment = adjusted_qty * best_match.equipment_cost

            line_items.append(CostLineItem(
                bim_type=bim_type,
                work_item=best_match,
                quantity=adjusted_qty,
                quantity_unit=best_match.unit,
                total_cost=total,
                labor_cost=labor,
                material_cost=material,
                equipment_cost=equipment,
                phase=phase,
                trade=._get_trade(category)
            ))

         line_items

     () -> [CostLineItem]:
        

        all_line_items = []

        
        grouped = qto_data.groupby([category_column, type_column]).agg({
            quantity_column: 
        }).reset_index()

         _, row  grouped.iterrows():
            items = .estimate_element(
                bim_type=row[type_column],
                category=row[category_column],
                quantity=row[quantity_column],
                quantity_unit=  
            )
            all_line_items.extend(items)

         all_line_items

     () -> :
        

        
        conversions = {
            (, ): ,
            (, ): ,
            (, ): ,
            (, ): ,
            (, ): ,
            (, ): ,
        }

        key = (from_unit.lower(), to_unit.lower())
        factor = conversions.get(key, )

         value * factor

     () -> :
        
        trade_map = {
            : ,
            : ,
            : ,
            : ,
            : ,
            : ,
            : ,
            : ,
            : 
        }
         trade_map.get(category, )

     () -> :
        

        
        records = []
         item  line_items:
            records.append({
                : item.bim_type,
                : item.work_item.description,
                : item.work_item.cwicr_code,
                : (item.quantity, ),
                : item.quantity_unit,
                : (item.work_item.unit_price, ),
                : (item.labor_cost, ),
                : (item.material_cost, ),
                : (item.equipment_cost, ),
                : (item.total_cost, ),
                : item.phase,
                : item.trade,
                : item.work_item.currency,
                : (item.work_item.confidence, )
            })

        df = pd.DataFrame(records)

        
        total_cost = df[].()
        total_labor = df[].()
        total_material = df[].()
        total_equipment = df[].()

        
        by_trade = df.groupby()[].().sort_values(ascending=)

        
        excel_path = 
         pd.ExcelWriter(excel_path, engine=)  writer:
            
            summary_data = {
                : [, , , ],
                : [total_cost, total_labor, total_material, total_equipment]
            }
            pd.DataFrame(summary_data).to_excel(writer, sheet_name=, index=)

            
            by_trade.to_frame().to_excel(writer, sheet_name=)

            
            df.to_excel(writer, sheet_name=, index=)

         {
            : excel_path,
            : total_cost,
            : total_labor,
            : total_material,
            : total_equipment,
            : by_trade.to_dict(),
            : (df),
            : line_items[].work_item.currency  line_items  
        }



 () -> :
    

     subprocess
     pathlib  Path

    
    ()
    subprocess.run([
        ,
        model_path,
        , 
    ])

    xlsx_path = Path(model_path).with_suffix()

    
    ()
    df = pd.read_excel(xlsx_path)

    
    estimator = BIMCostEstimator(
        qdrant_url=qdrant_url,
        language=language
    )

    
    ()
    line_items = estimator.estimate_from_qto(df)

    
    project_name = Path(model_path).stem
    result = estimator.generate_estimate_report(
        line_items=line_items,
        project_name=project_name,
        output_path=output_dir
    )

    ()
    ()
    ()

     result


 __name__ == :
    result = estimate_from_bim_model(
        model_path=,
        qdrant_url=,
        language=,
        output_dir=
    )

n8n Workflow

See: n8n_4_CAD_(BIM)_Cost_Estimation_Pipeline_4D_5D_with_DDC_CWICR.json

stages:
  - convert: RvtExporter → XLSX
  - detect_project: LLM identifies project type
  - generate_phases: LLM creates construction phases
  - decompose: LLM breaks types into work items
  - vector_search: Qdrant finds CWICR matches
  - calculate: Qty × Unit Price
  - validate: CTO review
  - report: HTML + Excel output

Output Example

╔══════════════════════════════════════════════════════════════╗
║                    COST ESTIMATE SUMMARY                      ║
║   Project: Office Building Berlin                             ║
║   Date: 2026-01-24                                           ║
╠══════════════════════════════════════════════════════════════╣

TOTAL PROJECT COST:                    EUR 4,523,678.00
───────────────────────────────────────────────────────────────
  Labor:                               EUR 1,847,234.00 (41%)
  Materials:                           EUR 2,312,456.00 (51%)
  Equipment:                           EUR   363,988.00 ( 8%)

BY TRADE
───────────────────────────────────────────────────────────────
  Concrete:                            EUR 1,234,567.00 (27%)
  Masonry:                             EUR   876,543.00 (19%)
  Steel Structure:                     EUR   654,321.00 (14%)
  MEP:                                 EUR   543,210.00 (12%)
  Finishes:                            EUR   432,109.00 (10%)
  Other:                               EUR   782,928.00 (18%)

CONFIDENCE ANALYSIS
───────────────────────────────────────────────────────────────
  High (>0.85):                        78%
  Medium (0.70-0.85):                  18%
  Low (<0.70):                          4%

╚══════════════════════════════════════════════════════════════╝

Resources

• CWICR Repository: https://github.com/datadrivenconstruction/OpenConstructionEstimate-DDC-CWICR
• Live Demo: https://openconstructionestimate.com
• Qdrant: https://qdrant.tech

---

"Resource-based costing separates physical quantities from volatile prices, enabling transparent and auditable estimates."