360◦ Inventory Management System: Design, Implementation, and Performance Analysis of a Real-Time Integrated Platform

360 Inventory Management System: Design, Implementation, and Performance Analysis of a Real-Time Integrated Platform

Shiwani Aggarwal

Assistant Professor CSE(Data Science)

Moradabad Institute of Technology, Moradabad, India shiwani.agarwal310@gmail.com

Dhruv Kumar Singh

Department of CSE(Data Science)

Moradabad Institute of Technology, Moradabad, India dt5541858@gmail.com

Tushar Sharma

Department of CSE(Data Science)

Moradabad Institute of Technology, Moradabad, India tusharsharma.tdau@gmail.com

AbstractModern enterprises require sophisticated inventory management solutions that transcend traditional stock tracking. This paper presents the comprehensive design, implementation, and evaluation of a 360 Inventory Management System (360- IMS), an integrated platform providing end-to-end visibility across procurement, storage, sales, and analytics. Unlike con- ventional systems, 360-IMS implements bidirectional real-time synchronization with e-commerce platforms, role-based access control, and predictive analytics capabilities. The system architec- ture leverages modern web technologies including React.js for the presentation layer, Spring Boot with RESTful APIs for business logic, and MySQL for persistent storage. Performance evaluation demonstrates 98.7% synchronization accuracy, 340ms average API response time, and 73% reduction in manual reconciliation effort compared to spreadsheet-based approaches. The system successfully handles concurrent operations from 500+ users while maintaining data consistency. Security analysis conrms robust JWT-based authentication and authorization mechanisms. This research contributes a practical, scalable framework suitable for small to medium enterprises seeking cost-effective inventory automation without the complexity of traditional ERP systems.

Index TermsInventory Management System, Real-Time Syn- chronization, RESTful API, E-Commerce Integration, React.js, Spring Boot, Stock Control Automation

1. INTRODUCTION

   Inventory management constitutes a critical operational component directly impacting organizational protability, cus- tomer satisfaction, and competitive positioning. Research in- dicates that inefcient inventory practices cost businesses approximately 8-12% of total sales revenue annually through stockouts, excess inventory, and operational inefciencies [1]. Traditional inventory management approachesmanual ledgers, spreadsheet-based tracking, and standalone desktop

   Anshuman Singh

   Department of CSE(Data Science)

   Moradabad Institute of Technology, Moradabad, India anshuman00002@gmail.com

   Avinash

   Department of CSE(Data Science)

   Moradabad Institute of Technology, Moradabad, India avinashyadav10112003@gmail.com

   applicationssuffer from fundamental limitations including lack of real-time visibility, susceptibility to human error, poor scalability, and inability to integrate with modern e- commerce ecosystems. While Enterprise Resource Planning (ERP) systems address many of these challenges, their high implementation costs (typically Rs. 1.2-6 Crore for SMEs), lengthy deployment cycles (6-18 months), and operational complexity render them impractical for small to medium enterprises.

   This research addresses the gap between inadequate legacy systems and cost-prohibitive enterprise solutions by presenting a 360 Inventory Management System. The term 360 signies complete lifecycle visibility encompassing product master data, real-time stock levels, multi-channel sales in-

   tegration, purchase order management, and business intelli- gence analytics. The system architecture prioritizes modularity, scalability, and integration capability while maintaining cost- effectiveness.

   Primary contributions of this research include: (1) architec- tural design of a modular inventory platform with bidirectional e-commerce synchronization, (2) implementation of REST- ful APIs with Spring Boot enabling exible data retrieval,

   (3) comprehensive performance benchmarking demonstrating production readiness, and (4) security analysis validating enterprise-grade access control mechanisms.

2. PROBLEM STATEMENT AND MOTIVATION

   1. Limitations of Current Approaches

      Small and medium enterprises face critical inventory man- agement challenges that existing solutions fail to adequately address:

      Real-Time Visibility Decit: Manual and spreadsheet- based systems require batch updates, creating information lag of 2-48 hours. This delay causes inventory discrepancies averaging 15-25% in multi-channel operations [2].

      Integration Fragmentation: E-commerce platforms, point- of-sale systems, and inventory databases often operate in isolation. Manual data transfer between systems introduces errors and consumes 8-15 hours weekly per business.

      Scalability Constraints: Spreadsheet solutions deteriorate signicantly beyond 500-1000 SKUs. Database corruption, formula errors, and le size limitations become critical bottle- necks.

      Analytics Inadequacy: Traditional systems provide histor- ical reporting but lack predictive capabilities. Decision-makers cannot identify trends, forecast demand, or optimize reorder points effectively.

      Cost-Complexity Imbalance: While ERP systems offer comprehensive functionality, implementation costs, training requirements, and ongoing maintenance create prohibitive bar- riers for SMEs.

   2. Research Objectives

   This project develops an inventory management platform addressing these limitations through: real-time bidirectional synchronization achieving ¡500ms latency; RESTful APIs en- abling seamless third-party integration; modular architecture supporting horizontal scaling to 10,000+ SKUs; role-based ac- cess control with JWT authentication; responsive web interface supporting desktop and mobile devices; deployment exibility (cloud, on-premises, or hybrid); and implementation cost ¡20% of comparable ERP solutions.

3. LITERATURE REVIEW

   A. Traditional Inventory Control Models

   Classical inventory optimization emerged from operations research in the mid-20th century. The Economic Order Quan- tity (EOQ) model, developed by Harris (1913), minimizes total inventory cost by balancing ordering and holding expenses. The EOQ formula:

   accuracy improving from 78% to 94% post-implementation [3]. However, ERP limitations include high total cost of own- ership (Rs.2-20 Crore for SMEs), complex customization re- quirements, lengthy implementation timelines (8-24 months), and user adoption challenges. Studies show 55-60% of ERP projects exceed budget or timeline constraints [4].

   1. Cloud-Based Inventory Platforms

      Software-as-a-Service (SaaS) inventory solutions emerged circa 2010, offering browser-based access, subscription pric- ing, and automatic updates. Platforms like TradeGecko (now QuickBooks Commerce), Cin7, and Zoho Inventory target SMEs with Rs.4,000-40,000 monthly pricing tiers. Research by Gartner (2024) projects cloud-based inventory manage- ment market growth at 14.2% CAGR through 2028 [5]. Key advantages include rapid deployment, predictable costs, and scalability.

   2. E-Commerce Inventory Integration

      E-commerce platforms (Shopify, WooCommerce, Magento) typically include basic inventory tracking. However, research identies critical gaps in multi-channel environments. A study by Harvard Business Review (2022) found 73% of retailers experienced inventory synchronization issues causing stock- outs or overselling [6]. API-based integration approaches using REST, GraphQL, or webhooks enable real-time synchroniza- tion with 2-5 second latecy.

   3. Modern Web Technologies

      Recent advances in web technologies enable sophisti- cated browser-based applications. React and Next.js provide component-based UI development with server-side rendering. Node.js enables JavaScript-based backend development with high concurrency through event-driven architecture. GraphQL addresses REST API limitations through exible query lan- guage. Research demonstrates GraphQL reduces data transfer by 30-50% compared to traditional REST endpoints [7].

   4. Research Gaps

   Literature analysis reveals unaddressed challenges: limited

   Q = 2DS

   H

   (1)

   research on cost-effective inventory solutions bridging manual systems and ERP complexity; insufcient empirical data on

   where D represents annual demand, S denotes ordering cost, and H signies holding cost per unit per year. Just-In-Time (JIT) inventory management, pioneered by Toyota Production System, minimizes inventory through demand-driven replen- ishment. ABC analysis categorizes inventory based on value contribution. While mathematically sound, these techniques require accurate, timely dataa challenge in manual systems.

   B. Enterprise Resource Planning Systems

   Modern ERP systems (SAP S/4HANA, Oracle NetSuite, Microsoft Dynamics) integrate inventory management with nance, procurement, manufacturing, and sales modules. Re- search by Panorama Consulting (2023) indicates ERP imple- mentations achieve 65% of anticipated benets, with inventory

   Spring Boot performance in inventory contexts; lack of open- source reference architectures for integrated platforms; and minimal comparative analysis of synchronization strategies using React-based e-commerce integrations. This research addresses these gaps through comprehensive system design, implementation, and evaluation.

4. CLASSIFICATION OF SYSTEMS

   Inventory management systems can be classied across multiple dimensions:

   By Deployment: On-premises systems operate within or- ganizational infrastructure. Cloud-based (SaaS) systems offer accessibility and scalability. Hybrid approaches combine both paradigms.

   TABLE I

   360 IMS Architecture

   HTTPS

   JDBC

   HTTP/S

   Integration

   E-commerce APIs Payment

   Data

   MySQL Redis Cache

   Application

   Spring Boot JWT

   REST Controllers Services

   Presentation

   React.js + Redux

   COMPREHENSIVE COMPARISON OF INVENTORY MANAGEMENT SYSTEMS

   System	Auto	RT	Cost	Scale	Integ
   Manual	V.Low	No	Rs. 0-40K	¡ 100	None
   Spreadsheet	Low	Part	Rs. 0-80K	¡500	Man
   Desktop	Med	Part	Rs. 1.6L-12L	¡2K	Ltd
   ERP	V.High	Yes	Rs. 1.2-20Cr	¿50K	Full
   SaaS	High	Yes	Rs. 50K-5L/yr	¡10K	API
   360-IMS	V.High	Yes	Rs. 4L-20L	¡15K	REST

   Fig. 1. Layered Architecture of 360 IMS

   By Integration: Standalone systems operate independently. Partially integrated systems connect with specic modules. Fully integrated platforms synchronize across procurement, sales, nance, and logistics.

   By Automation: Manual systems require human interven- tion. Semi-automated systems handle routine operations with oversight. Fully automated systems operate autonomously.

   By Architecture: Monolithic systems integrate functional- ity in single codebases. Service-oriented architectures decom- pose into discrete services. Microservices maximize modular- ity and independent scaling.

5. SYSTEM ARCHITECTURE

   1. Architectural Overview

      The 360 IMS implements a modern three-tier architecture optimized for scalability, maintainability, and performance. Figure 1 illustrates the layered design separating presentation,

      application, and data concerns. Table I compares different system categories.

   2. Technology Stack

      Frontend: React.js 18. React.js provides component-based architecture with virtual DOM for efcient rendering and updates. Hooks and functional components enhance code reusability and maintainability. Redux for state management ensures predictable application behavior. Component lifecycle optimization improves initial page load by 40-60%. TypeScript integration provides type safety reducing runtime errors by 15% [8].

      Backend: Spring Boot 3. Spring Boot provides enterprise- grade Java framework with embedded Tomcat server, auto- conguration, and production-ready features. Spring MVC handles HTTP requests efciently. Spring Data JPA simplies database operations with ORM capabilities. Built-in security features and dependency injection enhance maintainability.

      Spring Boots modular architecture supports 10,000+ concur- rent connections with thread pooling.

      API Layer: RESTful Architecture. REST APIs provide standardized, stateless communication using HTTP meth- ods (GET, POST, PUT, DELETE). JSON format ensures lightweight data exchange. Request/response model with proper HTTP status codes simplies error handling. Spring Boots @RestController annotations streamline endpoint de- velopment. API versioning and HATEOAS support future extensibility.

      Database: MySQL 8. MySQL selection criteria include: ACID compliance ensuring data consistency; InnoDB storage engine with row-level locking; B-tree and hash indexing optimizing query performance; support for transactions and foreign key constraints; horizontal scaling via master-slave replication; mature ecosystem with MySQL Workbench for administration. Benchmark testing shows MySQL handles 15,000+ transactions per second on standard hardware con- gurations.

      Caching: Redis. Redis provides in-memory data storage achieving ¡1ms read latency. Use cases include session man- agement, API response caching, and real-time inventory coun- ters. Implementing Redis caching reduces database load by 65-70% for read-heavy operations.

   3. Database Schema

      The normalized schema comprises 12 core tables: users, products, inventory, orders, order_items, suppliers, purchase_orders, stock_movements, locations, categories, sync_logs, and audit_logs. MySQL InnoDB engine provides ACID properties with row-level locking. Key indexing includes composite indexes on (product_id, location_id), B-tree indexes on timestamp elds, and full-text indexes on product descriptions using MySQLs FULLTEXT index capability.

   4. Security Architecture

   Authentication: JWT-based stateless authentication with 15-minute access tokens and 7-day refresh tokens. Tokens include encrypted user ID, role, and permissions.

   Authorization: RBAC implements four roles: Super Admin (full system access), Admin (inventory operations, reporting), Manager (order processing, viewing), and Staff (read-only access).

   Data Protection: AES-256 encryption at rest. TLS 1.3 in transit. PreparedStatement usage in JDBC prevents SQL injection. Input validation using Spring Boots validation anno- tations. Rate limiting (100 requests/minute per user) prevents abuse.

6. IMPLEMENTATION

   1. Real-Time Synchronization

      The bidirectional synchronization engine operates through webhook-based event propagation and React library integra- tions. E-commerce to 360-IMS ow: customer places order;

      Shopify triggers webhook; Sprig Boot controller validates payload; service layer creates order record; inventory service decrements stock using @Transactional annotation; analytics updates; conrmation sent within 500ms. 360-IMS to E- commerce ow: admin updates quantity in React interface; change triggers REST API call; Spring Boot service de- tects change; HTTP client (RestTemplate/WebClient) updates Shopify via their API; retry mechanism with @Retryable handles failures; service logs transaction in MySQL.

      E-commerce integration leverages React libraries including Axios for HTTP requests, custom hooks for API calls, and React Context for managing e-commerce connection state. The frontend communicates with Spring Boot REST endpoints which then interact with external e-commerce APIs using standard HTTP clients.

      Conict resolution implements last-write-wins with timestamp comparison. Discrepancies trigger automated rec- onciliation job using Springs @Scheduled annotation, com- paring system states and generating difference reports.

   2. Core Modules

      Product Management: Lifecycle management including creation, categorization, variants, pricing history, supplier link- ing. Bulk CSV import with validation. Barcode generation. CDN image delivery.

      Inventory Tracking: Real-time monitoring across loca- tions. Automatic reorder calculations using moving average demand. Low-stock email/SMS alerts. Stock transfers with approval workow. Cycle counting with discrepancy tracking. Order Processing: Multi-channel consolidation from e- commerce, POS, manual entry. Status tracking (pending, con- rmed, shipped, delivered, cancelled). Automated picking lists.

      Shipping label integration. Return processing.

      Analytics Dashboard: Real-time KPI visualization includ- ing turnover ratio, DSI, stockout frequency, gross margin. Historical trend analysis. ABC reports. Slow-moving identi- cation. Sales forecasting via linear regression.

   3. API Design

   RESTful API design follows industry best practices with resource-based URLs, proper HTTP methods, and standard- ized status codes. Spring Boot controllers expose endpoints for inventory operations. Example endpoints include: GET

   { }

   { }

   /api/products, POST /api/orders, PUT /api/inventory/ id , DELETE /api/products/ id . Request/response DTOs ensure type safety. Pagination implements page-based approach with RequestParam annotations handling 10,000+ records ef- ciently. Spring Boots built-in error handling provides con- sistent error responses. API documentation generated using SpringDoc OpenAPI (Swagger).

7. RESULTS AND PERFORMANCE

   1. Experimental Setup

      Infrastructure: Application Server (4-core CPU, 16GB RAM, SSD); Database Server (8-core CPU, 32GB RAM, NVMe SSD); 1Gbps LAN. Dataset: 5,000 SKUs, 50,000

      TABLE II

      API PERFORMANCE METRICS

      Operation	Avg	95th
      Product Search	145ms	320ms
      Inventory Query	89ms	180ms
      Order Creation	340ms	620ms
      Stock Update	52ms	110ms
      Dashboard	890ms	1450ms
      Report Gen	2100ms	3800ms
      E-comm Sync	420ms	780ms

      TABLE III

      OPERATIONAL EFFICIENCY COMPARISON

      Metric	Sheet	360-IMS	Gain
      Stock Update	8.5min	0.5min	94%
      Inventory Acc	82%	98.7%	+16.7pp
      Order Time	12min	3.2min	73%
      Reconcile/wk	6hrs	0.5hrs	92%
      Error Rate	15%	1.3%	91%

      transactions, 3 locations, 500 simulated users. Load testing via Apache JMeter with gradual ramp-up over 30 minutes executing realistic workows.

   2. Performance Benchmarks

      All critical operations maintain sub-second response times (Table II). Query optimization through indexing achieved 60- 75% improvement over unoptimized implementation.

   3. Synchronization Accuracy

      Testing 10,000 transactions over 30 days: Success Rate 98.7%; Average Latency 420ms; Failed Syncs 130 (1.3%); Retry Success 96.2%; Net Accuracy 99.95%. Failures: tran- sient network (45%), rate limiting (38%), invalid format (17%). Exponential backoff resolved 96.2% within 5 minutes.

   4. Comparative Analysis

      Results demonstrate substantial efciency gains (Table III). Reconciliation reduced from 6 hours to 30 minutes weekly. Accuracy improvement from 82% to 98.7% impacts customer satisfaction and reduces stockout costs.

   5. Scalability

      System handled: 500 concurrent users without degradation; 15,000 SKUs with sub-second queries; 50,000 daily transac- tions; 200 orders/minute peak throughput. Connection pooling (max 50) and Redis caching enabled efcient utilization. CPU peaked at 65%, memory at 72%, indicating substantial growth headroom.

   6. Cost-Benet Analysis

   360-IMS provides 88% savings versus ERP while delivering comparable SME functionality (Table IV). Higher upfront versus SaaS provides data ownership, customization, and elim- inates recurring fees.

   TABLE IV

   TOTAL COST OF OWNERSHIP (3-YEAR)

   Component	ERP	SaaS	360-IMS
   Implementation	Rs. 1.44Cr	Rs. 1.6L	Rs. 14.4L
   Licenses (3yr)	Rs. 72L	Rs. 12L	Rs. 0
   Infrastructure	Rs. 20L	Rs. 0	Rs. 9.6L
   Training	Rs. 16L	Rs. 2.4L	Rs. 4L
   Maintenance	Rs. 36L	Rs. 0	Rs. 6.4L
   Total	Rs. 2.88Cr	Rs. 16L	Rs. 34.4L

8. ADVANCED ANALYSIS

   1. Reorder Point Algorithm

      Dynamic reorder calculation using demand forecasting:

      ROP = (Davg × LT )+ SS (2)

      × ×

      where ROP = Reorder Point, Davg = Average daily demand (30-day moving average), LT = Lead time (days), SS = Safety stock = z D LT , z = Service level z-score (1.65 for 95%), D = Demand std deviation. Testing shows 68% stockout reduction versus xed points while maintaining 12% lower average inventory.

   2. Conict Resolution

      >

      { }

      For simultaneous updates, vector clock-based detection: each record maintains V = vims, vshopify incremented on updates. If VA VB, accept A. If VA VB (concurrent), apply rules: quantity uses minimum (prevent overselling); price uses latest timestamp; descriptive elds agged for review.

   3. Security Threat Model

   STRIDE analysis: Spoong (JWT signature, HTTPS pin- ning); Tampering (audit logs, versioning); Repudiation (im- mutable trail, crypto signing); Information Disclosure (eld encryption, role ltering); DoS (rate limiting, CAPTCHA, throttling); Privilege Escalation (least privilege, explicit checks). Penetration testing: zero critical, 3 medium issues (resolved).

9. DISCUSSION

   1. Key Findings

      The research validates hyotheses regarding modern in- ventory management. Multi-channel organizations experience 15-25% discrepancies without real-time sync. 420ms latency eliminates time-gap inconsistencies. Spring Boots efcient request handling and MySQLs optimized queries maintain consistent sub-second response times. Reacts component- based architecture with virtual DOM provides responsive user experience. The technology stack (React-Spring Boot- MySQL) optimal for 500-15,000 SKUs. User testing identied learning curve; contextual help reduced prociency time from 12 to 4 hours.

   2. Practical Implications

      For SMEs: prioritize real-time sync over features; evaluate total ownership cost; ensure data export capabilities; consider 3-5 year scalability; validate integration compatibility. For developers: event-driven architecture decouples components; GraphQL subscriptions enable elegant real-time patterns; com- prehensive logging essential; query optimization beats hard- ware upgrades; security-by-design beats security-by-addition.

   3. Limitations

   Limited manufacturing integration (retail/wholesale focus). Single-tenant architecture (multi-tenant requires modica- tions). Scale testing at 5,000 SKUs, 500 users (larger scale may need optimization). Single-region optimization (global requires edge caching, replication).

10. FUTURE RESEARCH

    1. AI Integration

       Demand Forecasting: ML models (ARIMA, LSTM) for prediction. Prophet shows 15-20% improvement over moving averages.

       Anomaly Detection: Unsupervised learning for irregular patterns indicating theft, spoilage, errors. Isolation Forest promising for outlier detection.

       Dynamic Pricing: Reinforcement learning for automated markdown optimization balancing turnover and margin.

    2. Advanced Automation

       Computer Vision: Deep learning for automated receiv- ing verication eliminating manual counting. MobileNet and YOLO suitable for real-time detection.

       RPA: Automate routine tasks: PO generation, supplier com- munication, invoice reconciliation.

       Voice Interface: NLP for hands-free queries/updates, valu- able in warehouses.

    3. Blockchain Integration

       Investigate distributed ledger for supply chain provenance. Smart contracts automate compliance, payments, quality work- ows. Hyperledger Fabric represents promising framework.

    4. IoT Sensors

       RFID/Barcode: Mobile scanning for cycle counting, receiv- ing. BLE beacons enable warehouse location tracking.

       Environmental: Temperature/humidity sensors for perish- ables, auto-agging at-risk inventory.

       Smart Shelving: Weight-based detection via load cells provides continuous visibility.

    5. Enhanced Analytics

       Prescriptive: Recommendation engines suggesting optimal policies, supplier selection, assortment planning.

       Network Optimization: Operations research algorithms (linear programming, network ow) for multi-location opti- mization.

       Customer Segmentation: Clustering algorithms identify purchase patterns enabling targeted positioning.

    6. Microservices Migration

       Refactor to containerized microservices using Docker and Kubernetes. Spring Boots modular design naturally supports microservices architecture. Benets include: independent scal- ing based on load patterns; technology stack exibility per service; improved fault isolation; simplied continuous de- ployment with Spring Cloud integration.

    7. Mobile-First Design

    Native applications (iOS/Android via React Native) opti- mized for warehouse staff, delivery, eld sales. PWA imple- mentation provides ofine functionality.

11. CONCLUSION

This research presented comprehensive design, implementa- tion, and evaluation of a 360 Inventory Management System addressing critical gaps in existing solutions. The system demonstrates that modern web technologiesReact.js for frontend, Spring Boot for backend, and MySQL for data

persistenceenable cost-effective, scalable platforms suitable for SMEs without sacricing functionality or performance.

Performance evaluation validates production readiness: 98.7% synchronization accuracy, sub-second response times, 73% reduction in manual reconciliation versus spreadsheets. TCO analysis indicates 88% savings versus traditional ERP while maintaining comparable SME functionality.

The modular architecture leveraging Spring Boots enter- prise features, RESTful API design, and Reacts component- based UI provides extensibility foundation. Integration with e-commerce platforms through React libraries and REST end- points demonstrates practical applicability. Future directions including AI integration, blockchain provenance, and IoT connectivity promise further improvements.

This work contributes practical reference architecture for developers building integrated inventory platforms using React-Spring Boot-MySQL stack and provides empirical performance data informing organizational technology de-

cisions. The 360 approachemphasizing complete lifecy- cle visibility, real-time synchronization, and seamless inte-

grationrepresents effective paradigm for modern inventory management in digitally-driven business environments.

REFERENCES

1. K. C. Laudon and J. P. Laudon, Management Information Systems: Managing the Digital Firm, 17th ed. Pearson, 2022.

2. D. Simchi-Levi, P. Kaminsky, and E. Simchi-Levi, Designing and Managing the Supply Chain, 4th ed. McGraw-Hill, 2021.

3. Panorama Consulting Solutions, 2023 ERP Report, Denver, CO, 2023.

4. R. S. Pressman and B. R. Maxim, Software Engineering: A Practi- tioners Approach, 9th ed. McGraw-Hill, 2020.

5. Gartner Inc., Market Guide for Inventory Management Systems, Stamford, CT, Rep. G00760845, 2024.

6. D. Bell, S. Gallino, and A. Moreno, Inventory Showrooms and Cus- tomer Migration in Omnichannel Retail, Harvard Business Review, vol. 100, no. 4, pp. 86-94, 2022.

7. B. V. Belli, K. Lautenbach, and M. Andrews, Performance Analysis of GraphQL and REST, in Proc. IEEE Int. Conf. Web Services, 2021, pp. 145-152.

8. G. Gao, et al., To Type or Not to Type: Quantifying Detectable Bugs in JavaScript, in Proc. 39th Int. Conf. Software Engineering, 2017, pp. 758-769.