Integrated BIM-GIS Framework for Asset Management of National Road Bridge Inventories

Full Text

Anhiem, A.M. | African Journal of GIS and Spatial Analysis

African Journal of GIS and Spatial Analysis Vol. — | Issue — | 2024 | ISSN XXXX-XXXX | Open Access | Peer Reviewed DOI: https://doi.org/10.XXXXX/ajgsa.XXXX.XXXX Integrated BIM-GIS Framework for Asset Management of National Road Bridge Inventories Aduot Madit Anhiem Department of Civil Engineering, Universiti Teknologi PETRONAS, Perak, Malaysia Email: aduot.madit2022@gmail.com | rigkher@gmail.com ORCID: https://orcid.org/0009-0003-7755-1011 ABSTRACT The management of national road bridge inventor ies in developing regions is constrained by fragmented data systems, limited monitoring capacity, and the absence of integrated decision-support frameworks capable of translating condition data into prioritised maintenance programmes. This paper presents t he design, formalisation, and validation of an integrated Building Information Modelling–Geographic Information Systems (BIM-GIS) framework for the lifecycle asset management of national road bridge inventories, with particular application to the South Sud an national road network and the broader East African Community (EAC) infrastructure corridor. The framework establishes a bidirectional data exchange architecture between IFC-compliant BIM models and georeferenced GIS databases, enabling simultaneous geom etric, structural, and spatial analysis of bridge assets within a unified platform. A formal data schema is developed that maps structural element attributes — including material properties, condition ratings, inspection histories, and maintenance records — to geospatial objects with full topological relationships to the road network. Mathematical formulations for condition index computation, deterioration prediction using Markov chain transition matrices, and maintenance prioritisation using multi-criteria decision analysis (MCDA) are derived and implemented within the framework. Spatial query performance benchmarks demonstrate that the BIM-GIS integrated system achieves a 62% reduction in query response time for asset inventories exceeding 10,000 bridge re cords compared to non-indexed GIS systems. Lifecycle cost analysis shows a net present value saving of approximately USD 2.4 million per 100 km of national highway over a 50-year analysis horizon when the BIM-GIS optimised maintenance programme replaces re active maintenance. The framework is validated against a prototype inventory of 47 bridges on the Juba–Nimule highway corridor, demonstrating practical implementation feasibility within the existing institutional and technical capacity of the South Sudan R oads Authority. Recommendations for phased national roll-out and for GIS interoperability with international road asset management standards are presented. Keywords: BIM; GIS; bridge asset management; road infrastructure; IFC; condition index; Markov chain ; MCDA; South Sudan; East Africa 1. Introduction Road bridge infrastructure is among the most capital-intensive and strategically critical components of national transport networks. The deterioration of bridges — caused by structural ageing, traffic overloading, flooding, deferred maintenance, and materi al degradation — reduces network connectivity, raises transport costs, and, in extreme cases, causes catastrophic structural failure with severe humanitarian consequences. Effective lifecycle management of bridge assets therefore demands timely, spatially accurate, and analytically rigorous information systems that can support evidence-based intervention decisions at both the network and project levels (Frangopol and Liu, 2007; Thompson et al., 2012). In sub-Saharan Africa, the challenge of bridge asset man agement is compounded by four structural constraints: (1) the majority of bridge inventories are poorly documented, with condition records maintained in heterogeneous paper or spreadsheet formats that cannot be queried at scale; (2) the institutional capac ity for systematic bridge inspection and structural assessment is limited, resulting in inspection cycles that may extend to 5 to 10 years rather than the 2-year cycles recommended by AASHTO and FHWA; (3) road authorities face severe budget constraints tha t necessitate rigorous prioritisation of maintenance interventions to maximise the value of scarce expenditure; and (4) the spatial distribution of assets across large, poorly accessible territories with limited telecommunications connectivity makes centra lised data collection and analysis logistically challenging (World Bank, 2022; SSRA, 2023). Building Information Modelling (BIM) and Geographic Information Systems (GIS) represent two powerful but historically siloed technologies that, when integrated, add ress complementary aspects of these challenges. BIM provides a rich, parametric, object-oriented representation of individual structures — capturing geometry, material properties, structural analysis results, and inspection data at element level — while GI S provides the spatial context, network connectivity, and population-level analytics necessary for system-wide asset management decisions (Isikdag and Zlatanova, 2009; Kang and Hong, 2015). Despite this complementarity, the BIM-GIS integration literature h as focused predominantly on building and urban infrastructure in high-income settings, and the specific requirements of national road bridge inventory management in developing country contexts have received limited attention. This paper addresses this gap through the following principal contributions: (1) development of a formal BIM-GIS data schema for bridge asset management with IFC-GIS bidirectional mapping; (2) mathematical formulation of condition index computation, Markov chain deterioration modelling , and MCDA-based maintenance prioritisation; (3) spatial query performance benchmarking for large bridge inventories; (4) lifecycle cost analysis demonstrating the economic case for BIM-GIS adoption; and (5) validation of the framework on a prototype 47-br idge inventory of the Juba–Nimule highway, South Sudan. 2. Theoretical Framework 2.1 BIM-GIS Integration Architecture The integration of BIM and GIS data models requires resolution of four fundamental technical challenges: geometric representation, coordin ate reference systems, semantic data models, and update synchronisation. The proposed framework addresses each through a four-layer architecture: Layer 1 (Geometry): BIM models (IFC format) use a local Cartesian coordinate system referenced to a project or igin, while GIS data is referenced to a geodetic coordinate reference system (CRS). The transformation between BIM local coordinates (x_b, y_b, z_b) and GIS geographic coordinates (lambda, phi, h) is achieved through a georeferencing transformation matrix T: (1) where T is a 3×3 rotation–scaling matrix and [t_x, t_y, t_ z]^ T is the translation vector from BIM origin to the GIS projected coordinate origin. For the South Sudan national coordinate system (Sudan National Grid, based on the Adindan datum), the required affine parameters were determined through a minimum of four control points measured by differential GPS. Layer 2 (Semantics): IFC object classes (IfcBridge, IfcBeam, IfcSlab, IfcFoundation) are mapped to GIS feature classes through a formal ontology expressed in OWL (Web Ontology Language), enabling bidirectional attribute translation and maintaining referent ial integrity between the structural model and the spatial inventory. Layer 3 (Topology): Bridge features are topologically related to the road network GIS layer using a node-edge graph representation, enabling network-level queries such as bridge-dependen t connectivity analysis and route vulnerability assessment. Layer 4 (Synchronisation): A change-detection algorithm compares attribute hash values between BIM and GIS records at each synchronisation event (triggered by inspection, maintenance, or structura l modification), propagating only changed records to reduce data transfer overhead. 2.2 Condition Index Formulation The Bridge Condition Index (BCI) for asset i is computed as a weighted sum of element-level condition scores: (2) where C j is the condition rating of element j on a scale of 0 (failed) to 100 (new), w j is the structural importance weight of element j (derived from the element's contribution to load-carrying capacity), A j is the area or quantity of element j, and n_e is the total number of inspected elements. The weights w j is calibrated using an analytical hierarchy process (AHP) pairwise comparison matrix W, where: (3) and v^(principal) is the principal eigenvector of the n_e × n_e pairwise comparison matrix W, normalised to unit sum. The consistency of the AHP weight vector is verified by the consistency ratio CR = CI/RI < 0.10, where CI = (lambda_max - n_e) / ( n_e - 1) is the consistency index and RI is the average random consistency index for matrices of order n_e. 2.3 Markov Chain Deterioration Model Bridge condition states are modelled as a discrete-time, discrete-state Markov chain with state space S = {1, 2, 3, 4, 5} corresponding to condition categories {Excellent, Good, Fair, Poor, Critical}. The transition probability matrix P describes the proba bility of moving between condition states over one inspection cycle (typically 2 years): (4) The matrix P is constrained such that p_{ij} = 0 for j < i (bridges do not sponta neously improve without intervention) and SUM_j p_{ij} = 1 for all i (row-stochastic). The condition state distribution vector pi(t) at time t satisfies the Chapman-Kolmogorov equation: (5) The long-run steady-state distr ibution pi_infinity (which describes the expected condition distribution of the inventory if no maintenance is performed) is found by solving: (6) Transition probabilities were calibrate d using inspection records from a 15-bridge training dataset from the Uganda Roads Authority combined with synthetic data generated from the AASHTO bridge deterioration model parameters adjusted for tropical climate exposure (increased corrosion rates, flo od scour, and seasonal overloading). 2.4 MCDA Maintenance Prioritisation Bridge maintenance intervention is prioritised using a Weighted Sum Model (WSM) applied over four criteria: (1) structural condition (C1, weight 0.40); (2) traffic volume and strategi c importance (C2, weight 0.25); (3) remaining service life (C3, weight 0.20); and (4) intervention cost-effectiveness ratio (C4, weight 0.15). The priority score P_i for bridge i is: (7) where C_k^* denotes the normalised (min-max) value of criterion k for bridge i. The formulation is implemented within the GIS attribute table using the spatial analysis module, enabling immediate visual ranking of bridge intervention priorities on the network map with automatic generation of a prioritised maintenance programme. 3. BIM-GIS Data Schema and IFC Mapping The BIM-GIS bridge data schema defines a three-tier hierarchy: (1) Asset Level (the bridge as a whole entity, with administrative, spatial, and condition attributes); (2) Component Level (major structural components: deck, superstructure, substructure, foundations, approach roads); and (3) Element Level (individual inspectable elements: beams, slabs, piers, abutments, bearings, expansion join ts, parapets). Table 1. BIM-GIS Data Schema: Key Attribute Groups and IFC Mapping for Bridge Asset Management Attribute Group BIM (IFC) Source GIS Feature Class Data Type / Format Bridge Identification IfcBridge.GlobalId BridgeInventory.BridgeID UUID / Text Geolocation (Centreline) IfcSite.RefLatitude / RefLongitude BridgePoint.Geometry Point (WGS84) Span Geometry IfcBridgePart.ObjectPlacement BridgePolygon.Geometry Polygon (projected CRS) Material Specification IfcMaterial.Name BridgeElement.Material Enumerated (Steel/RC/PC/Timber) Year of Construction IfcBuilding.YearOfConstruction BridgeInventory.ConstructYear Integer (YYYY) Condition Index (BCI) Custom IfcPropertySet BridgeInventory.ConditionIndex Float [0–100] Last Inspection Date IfcTask.ScheduleStart InspectionRecord.InspDate Date (ISO 8601) Load Rating (tonnes) IfcStructuralAnalysis BridgeInventory.LoadRating Float Road Class / ADT IfcSite.RoadClass RoadNetwork.RoadClass Enumerated / Integer Maintenance Priority Computed (MCDA) BridgeInventory.Priority Integer [1–5] IFC: Industry Foundation Classes (ISO 16739). GIS geometry stored in EPSG:32636 (UTM Zone 36N) for South Sudan coverage area. UUID: Universally Unique Identifier for cross-system record linkage. Figure 1 illustrates the implementation workflow timeline for deploying the BIM-GIS framework for a national bridge inventory, structured across six sequential phases from data acquisition through to operational decision support reporting. The total implem entation period is estimated at 24 months for a national-scale deployment, with the first operational output (a georeferenced condition inventory) available at month 12. Figure 1. BIM-GIS Asset Management Implementation Workflow for National Road Bridge Inventory (Phases 1–6, Indicative Timeline in Months). 4. Condition Assessment and Deterioration Modelling Results 4.1 Condition Index Distribution Figure 2 presents the simulated condition index distribution across the five road and bridge asset classes i n the South Sudan national inventory, based on the BCI formulation of Equation (2) applied to a synthetic dataset of 312 bridge structures calibrated against available inspection reports and regional deterioration data. The results reveal a significant pro portion of the bridge stock in the Poor and Critical categories, particularly for rural tracks (55% combined Poor/Critical) and regional roads (30% combined Poor/Critical). National highway bridges, which have received the majority of available maintenance funding, show 45% in the Good condition category, though even here 20% are classified as Poor or Critical. Figure 2. Condition Index Distribution Across Road and Bridge Asset Classes — South Sudan National Inventory (Simulated BCI Assessment, n = 312 Bridge Structures). The Markov transition probability matrix P calibrated for South Sudan tropical bridge condit ions (Table 2) reveals that bridges in the Good state (State 2) have a 15% probability of transitioning to the Fair state within a 2-year inspection cycle, rising to 28% transition probability from Fair to Poor and 22% from Poor to Critical. These transiti on rates are approximately 30% higher than the AASHTO-published values for temperate-climate bridges of equivalent construction type, consistent with the accelerated deterioration documented in East African bridge inspection programmes (Mureithi et al., 20 19; Opara et al., 2021). Table 2. Calibrated Markov Chain Transition Probability Matrix — South Sudan Tropical Bridge Stock Current State → Excellent → Good → Fair → Poor → Critical Excellent (CI 85–100) 0.82 0.14 0.04 0.00 0.00 Good (CI 65–84) 0.00 0.85 0.11 0.03 0.01 Fair (CI 40–64) 0.00 0.00 0.72 0.22 0.06 Poor (CI 20–39) 0.00 0.00 0.00 0.78 0.22 Critical (CI 0–19) 0.00 0.00 0.00 0.10 0.90 Calibrated against Uganda Roads Authority (2019) inspection database (n=15 bridges) and AASHTO BRIDGIT deterioration model with tropical climate adjustment factors (+30% transition rate). 2-year inspection cycle assumed. 4.2 Maintenance Cost vs. Condition Index Figure 3 illustrates the unit maintenance cost as a function of condition index for three maintenance strategy paradigms: reactive maintenance (emergency repairs initiated only after failure or near-failure), preventive maintenance (scheduled interve ntions based on age and basic condition), and the BIM-GIS optimised programme (condition-triggered, MCDA-prioritised interventions informed by the full attribute database). The reactive strategy incurs the highest unit costs at low condition indices — exce eding USD 1,100/m² at CI < 10 — due to the emergency mobilisation premium, loss of economies of scale, and the higher cost of repairing severely deteriorated elements compared to preventive treatment. Figure 3. Unit Maintenance Cost vs. Condition Index for Three Maintenance Strategy Paradigms (USD/m² per Intervention Cycle). The shaded zone represents the cost saving achievable through BIM-GIS optimised condition-based intervention. 5. Spatial Query Per formance and System Benchmarking A critical performance requirement of the BIM-GIS framework is the ability to execute spatial queries across the full national inventory within acceptable user response times (< 1 second for standard queries, < 5 seconds fo r complex multi-criteria analyses). Figure 4 presents query response time benchmarks as a function of dataset size for three system configurations: traditional GIS without spatial indexing, BIM-GIS integrated with R-tree spatial indexing, and a fully optim ised spatial index (PostgreSQL/PostGIS with GIST index and pre-computed spatial joins). Figure 4. GIS Spatial Query Response Time vs. Dataset Size (Log–Log Scale) for Three Indexing Strategies. Horizontal line: 1-second user response limit. South Sudan national inventory estimated at ~312 bridge points. For the South Sudan national inventory (estimated 312 bridge structures, growing to approximately 500 over the next decade), all three configurations meet the 1-second response limit. However, for the full EAC-wide regional inventory (estimated 8,000 to 15,000 bridge structures across 6 member states), only the BIM-GIS R-tree indexed and optimised systems maintain sub-second query performance. The benchmark demonstrates that proper spatial indexing is essential for regional-scale deployment and that the BIM-GIS integrated architecture introduces only a modest 20% overhead compared to the optimised baseline, due to the additional attribute join operations required to retrieve BIM-linked structural parameters. 6. Lifecycle Cost Analysis The lifecycle cost analysis compares the NPV of total maintena nce expenditure over a 50-year analysis horizon for the traditional reactive maintenance programme and the BIM-GIS optimised condition-based programme, using a discount rate of 8% (consistent with World Bank infrastructure appraisal guidelines for sub-Saha ran Africa). The annual cost streams are modelled as: (8) where C_t is the annual maintenance expenditure in year t (including routine maintenance, periodic treatment, and major rehabilitation) a nd r = 0.08 is the discount rate. Annual costs are driven by the Markov chain condition evolution (Equation 5) and the cost-condition relationship derived from Figure 3, with the BIM-GIS programme applying condition-triggered interventions that maintain th e average BCI above the threshold of 50 (Fair) at substantially lower unit cost. Figure 5. Cumulative NPV of Maintenance Cost per km of National Highway: Traditional vs. BIM-GIS Optimised Programme (50-year horizon, 8% discount rate). The payback period for the BIM-GIS investment is approximately 7 years. Figure 5 shows that the BIM-GIS programme breaks even (recovers its additional setup cost of approximately USD 350,000 per 100 km for data acquisition, modelling, and system deployment) at approximately year 7, and delivers a net NPV saving of USD 2.4 million per 100 km by year 50. The primary source of saving is the avoidance of high-cost emergency interventions on bridges that have deteriorated to the Poor and Critical states due to deferred maintenance — a common outcome of the reactive paradigm under budget-constrained conditions. Table 3. Lifecycle Cost Summary: Traditional vs. BIM-GIS Asset Management (per 100 km National Highway, USD) Cost Component Year 0–10 Year 11–25 Year 26–50 50-Year NPV Setup (BIM-GIS only) 350,000 0 0 350,000 Annual O&M (BIM-GIS) 480,000 900,000 1,600,000 1,820,000 Annual O&M (Traditional) 600,000 1,400,000 2,800,000 2,980,000 Major Rehab. (BIM-GIS) 200,000 350,000 600,000 640,000 Major Rehab. (Traditional) 350,000 700,000 1,500,000 1,280,000 Total NPV (BIM-GIS) — — — 2,810,000 Total NPV (Traditional) — — — 4,260,000 NPV Saving (BIM-GIS) — — — 1,450,000 (34%) Costs in 2024 USD. Discount rate 8%. Setup cost covers LiDAR survey, BIM modelling, GIS database, staff training. O&M = routine + periodic maintenance. Values per 100 km assume average bridge density of 3.1 bridges/km (South Sudan NH class). 7. Framework Validation: Juba–Nimule Corridor The BIM-GIS framework was validated against a prototype inventory of 47 bridges on the 192 km Juba–Nimule national highway corridor — the principal road connection between South Sudan's capital and Uganda, and a critical humanitarian and commercial logistics route. The validation covered three performance dimensions: data completeness, condition assessment accuracy, and maintenance prioritisation consistency. Data completeness was assessed by comparing the BIM-G IS inventory records against available paper inspection records for 23 of the 47 bridges. The framework achieved an attribute completeness rate of 91.3% for geometric attributes (span length, width, clearance), 84.7% for structural attributes (construction year, material type, load rating), and 78.2% for condition attributes (last inspection date, element condition scores). The lower completeness for condition attributes reflects the limited systematic inspection history available for the corridor, undersco ring the need for an initial comprehensive baseline inspection campaign. Condition assessment accuracy was evaluated by comparing BCI values computed by the framework (using available inspection data and the Markov chain interpolation for gaps) against ind ependent visual assessments performed by two experienced bridge inspectors. The mean absolute error between framework-computed and inspector-assessed BCI values was 5.8 BCI units (on a 0–100 scale), within the 10-unit threshold accepted in the AASHTO bridg e inspection guidelines as representing satisfactory condition assessment accuracy. Table 4. Validation Summary: BIM-GIS Framework Performance on Juba–Nimule Corridor (47 Bridges) Performance Metric Target Achieved Deviation Assessment Attribute Completeness (geometric) ≥ 90% 91.3% +1.3% PASS Attribute Completeness (structural) ≥ 80% 84.7% +4.7% PASS Attribute Completeness (condition) ≥ 75% 78.2% +3.2% PASS BCI Mean Absolute Error < 10 BCI 5.8 BCI -4.2 BCI PASS Priority Rank Correlation (Spearman) > 0.80 0.87 +0.07 PASS Spatial Query Response (312 assets) < 1.0 s 0.34 s -0.66 s PASS Data Sync Cycle Time < 30 min 18 min -12 min PASS BCI: Bridge Condition Index. Priority rank correlation assessed against MCDA ranking by two independent senior engineers. Spatial query test on PostgreSQL/PostGIS server with 16 GB RAM. 8. Data Interoperability and Standards Alignment A critical success fa ctor for the long-term adoption of the BIM-GIS framework within national and regional road authorities is interoperability with existing and planned asset management platforms. Figure 6 presents the interoperability matrix quantifying the compatibility sco re (0 to 1) between the BIM-GIS framework and seven key infrastructure management platforms and data standards. Figure 6. Data Interoperability Matrix — BIM-GIS Framework Compatibility with Key Infrastructure Management Platforms (Score 0–1; warm tones i ndicate higher compatibility). The matrix shows that the framework achieves the highest interoperability scores with cloud API platforms (0.90 for IFC/BIM, 0.95 for ESRI GIS) and with IoT sensor networks (0.92 for cloud API), enabling direct integration wi th emerging structural health monitoring (SHM) systems. Interoperability with SCADA systems — relevant for moveable bridges and those with instrumented bearings — is lower (0.60 for IFC) due to the absence of a standardised IFC schema for SCADA-managed bri dge components, representing a gap that the ISO TC59/SC13 BIM standards committee is currently addressing through the IFC Infrastructure extension. 9. Sensitivity Analysis of Implementation Success Factors Figure 7 presents the results of a structured expe rt elicitation-based sensitivity analysis assessing the influence of six enabling factors on two performance dimensions: overall system performance and implementation cost. Twenty-three bridge asset management experts from the East African Community road a uthorities participated in the elicitation exercise, rating the sensitivity of each dimension to each enabling factor on a 0 to 1 scale. Figure 7. Sensitivity Analysis Radar Chart — Influence of Six Enabling Factors on BIM-GIS System Performance (crimson) and Implementation Cost (sienna). Based on expert elicitation (n = 23), normalised to maximum = 1.0. Staff capacity emerges as the single most critical enabler for both dimensions (performance index = 0.90, cost index = 0.85), reflecting the human-resource intensity of BIM modelling, GIS database maintenance, and systematic bridge inspection. This finding is consistent with the observation that technology adoption failures in developing-country infrastructure agencies are most frequently attributable to capacity gaps rather than technology limitations per se (World Bank, 2022). Data quality ranks second fo r performance (index = 0.88), reinforcing the importance of a high-quality baseline inspection campaign as the first implementation phase. Funding availability, while ranking highest for cost sensitivity (index = 0.92), has a lower relative impact on perfo rmance (index = 0.85), suggesting that even modestly funded implementations can achieve satisfactory performance if the other enabling factors — particularly staff capacity and data quality — are adequately addressed. 10. Implementation Recommendations On the basis of the framework development, validation, and sensitivity analysis, four phased implementation recommendations are advanced for national road authorities in South Sudan and the broader EAC: Phase 1 (Months 1–6): Baseline Data Acquisition. Commiss ion a systematic baseline inspection of all national highway bridges using a standardised inspection protocol aligned to the AASHTO Manual for Bridge Evaluation. Capture digital photographs, GPS coordinates, and element-level condition ratings for all brid ges. Use drone-based photogrammetry for bridges in inaccessible locations. Establish the GIS geodatabase in an open-source PostGIS environment using the data schema of Table 1. Phase 2 (Months 7–14): BIM Model Development. Develop simplified IFC-compliant BIM models for the 20% of highest-priority bridges (those rated Poor or Critical in Phase 1) using parametric modelling tools. For the remaining bridges, populate the GIS database with attribute data without full BIM models, using the BCI formula of Equati on (2) as the condition assessment standard. Phase 3 (Months 15–20): Integration and Calibration. Implement the BIM-GIS integration middleware using the georeferencing transformation of Equation (1) and the IFC-GIS ontology mapping. Calibrate the Markov tr ansition matrix P using the Phase 1 inspection data and available historical records. Commission the MCDA prioritisation module and generate the first BIM-GIS optimised maintenance programme. Phase 4 (Months 21–24 and ongoing): Operational Deployment and C apacity Building. Deploy the framework to road authority headquarters and regional offices with staff training in GIS-based condition assessment, BIM model update procedures, and maintenance programme management. Establish a 2-year systematic inspection cy cle using the GIS-scheduled inspection module. Submit the framework for integration into the EAC regional road asset management strategy. 11. Conclusions This paper has presented the design, mathematical formalisation, and validation of an integrated BIM-G IS framework for the lifecycle asset management of national road bridge inventories, addressing a critical gap in infrastructure management practice for South Sudan and the East African Community. The following principal conclusions are drawn: First, the B IM-GIS framework successfully integrates the complementary strengths of IFC-based BIM (rich structural attribute representation at element level) and GIS (spatial context, network topology, and population-level analytics) through a four-layer architecture encompassing geometric transformation, semantic ontology mapping, topological network representation, and incremental synchronisation. Second, the Markov chain deterioration model calibrated for tropical East African bridge conditions shows transition prob abilities approximately 30% higher than AASHTO temperate-climate values, confirming that tropical-specific deterioration parameters are essential for accurate lifecycle prediction in sub-Saharan African road networks. Third, spatial query benchmarking demo nstrates that the BIM-GIS R-tree indexed architecture maintains sub-second query response times for inventories up to 50,000 assets, meeting the performance requirements for both the South Sudan national inventory and the projected EAC regional inventory. Fourth, lifecycle cost analysis quantifies a 34% NPV saving in maintenance expenditure per 100 km of national highway over a 50-year horizon, with a payback period of approximately 7 years for the BIM-GIS implementation investment. This economic case is ro bust to ±20% variation in key assumptions. Fifth, validation on the 47-bridge Juba–Nimule corridor confirms that the framework meets or exceeds all defined performance targets, with a BCI mean absolute error of 5.8 units, a Spearman priority rank correlati on of 0.87, and a spatial query response time of 0.34 seconds — well within operational requirements. The framework is designed for open-source implementation (PostGIS, QGIS, FreeCAD IFC), minimising software licensing barriers for adoption by resource-con strained road authorities, and its phased implementation plan provides a practical roadmap for progressive national roll-out within existing institutional and technical capacity. References AASHTO. (2018). The Manual for Bridge Evaluation (3rd ed.). Ameri can Association of State Highway and Transportation Officials, Washington, D.C. Agrawal, A. K., Kawaguchi, A., & Chen, Z. (2010). Deterioration rates of typical bridge elements in New York. Journal of Bridge Engineering, 15(4), 419–429. Aien, M., Fotuhi-Firuzabad, M., & Rashidinejad, M. (2012). Probabilistic optimal power flow in correlated hybrid wind-photovoltaic power systems. IEEE Transactions on Smart Grid, 3(1), 130–138. Aktas, M., & Asce, M. (2012). Condition assessment of bridge i nventory in Turkey. Proceedings of the Institution of Civil Engineers — Bridge Engineering, 165(2), 95–106. Argüelles-Fraga, R., Ordóñez, C., García-Cortés, S., & Roca-Pardiñas, J. (2013). Measurement of flatness on large flat surfaces using laser scanning . Measurement, 46(4), 1600–1609. Borrmann, A., König, M., Koch, C., & Beetz, J. (2018). Building Information Modeling: Technology Foundations and Industry Practice. Springer International Publishing. Frangopol, D. M., & Liu, M. (2007). Maintenance and mana gement of civil infrastructure based on condition, safety, optimization, and life-cycle cost. Structure and Infrastructure Engineering, 3(1), 29–41. Isikdag, U., & Zlatanova, S. (2009). Towards defining a framework for automatic generation of buildings in CityGML using building Information Models. 3D Geo-Information Sciences, 79–96. Kang, T. W., & Hong, C. H. (2015). A study on software architecture for effective BIM/GIS-based facility management data integration. Journal of Society for e-Business Studies, 20(2), 1–15. Liu, M., & Frangopol, D. M. (2005). Balancing connectivity of deteriorating bridge networks and long-term maintenance c