Full Text

Research Article

Microservices Architecture for Optimizing Warehouse Management Systems in Global Supply Chains


1. Abstract

In today's globalized economy, where logistics and supply chain chains are getting increasingly global in nature, the efficiency and scalability of a warehouse management system become very important. Traditional monolithic architectures for WMS can have difficulty handling such dynamic and high-volume operations across diverse geographical locations. This paper proposes a microservices-based architecture that decomposes traditional WMS functionalities into domain-specific microservices, thereby enhancing scalability, resilience and efficiency. Therefore, it has been divided into two major environments: Origin Country and Destination Country. This is supported by dedicated microservices: Core Services, Inbound Services and Outbound Services that work together to provide support for a full product lifecycle, starting from the creation of the product until its final delivery. Later, the architecture is then validated in detail through a flow diagram, depicting interactions between microservices in different environments, modularity and fault tolerance of the system.

Keywords:
Supply Chain, Warehouse Management System, E-Commerce, Micro Services, SOA
2. Literature Review
The rapid evolution in global supply chains and e-commerce has placed new, unprecedented demands on warehouse management systems. Traditional WMS architectures, initially developed for the storage and movement of inventory within warehouse facilities, usually cannot keep pace with supply chain complexities in today's world. The literature review performed a study on the existing architectures of WMS, highlighting challenges that traditional systems had to bear, introducing benefits given by microservices in logistics management.
3. Overview of Existing WMS Architectures
Traditional WMS architectures were traditionally designed to be monolithic systems - that is, all the core functionalities, ranging from inventory management to the processing of orders and scheduling shipment are hardwired into one application. The structure makes deployment and management easy but has huge limitations concerning flexibility, scalability and maintenance. General WMS architecture is a composition of various modules integrated together, catering to operational activities ranging from order receipt, inventory management, picking, packing and shipping.

In the case of legacy WMS systems, these modules normally communicate via tightly coupled integrations through direct database connections or shared in-memory data structures. This architecture works for small-scale operations but fails to scale up when it is required to manage large warehouses or multi-warehouse networks
15. Traditional WMS implementations are built around a centralized database that stores all operational data. In fact, any scaling effort-through the addition of new modules or increased concurrency transactions-requires heavy loads on this central data repository which might lead to performance bottlenecks and eventual failures of such systems under heavy demand.

In order to overcome such limitations, some newer generations of WMSes have embraced a Service-Oriented Architecture
1, whereby services are decoupled and interact via standardized communication protocols like REST or SOAP. This can indeed provide a certain degree of modularity, but the reliance on shared services will still affect fault tolerance and scalability, especially for integrations with external systems.
4. Challenges in Traditional WMS Implementations
At the core of supply chain logistics, traditional WMS solutions have a number of disadvantages inherent in their structure, which severely restrain their power and agility for modern, high-volume operations. Among these factors is scalability. As companies grow and warehouse operations expand, the monolithic architecture of a traditional WMS inhibits scaling of particular functions independently. For example, if there is an increase in the volume of order processing during peak seasons, such as Black Friday or Cyber Monday, the system may slow down or even crash since the WMS cannot handle the additional load on its tightly coupled modules.

Another critical challenge is system resilience or fault tolerance. In a typical WMS, the failure of a module at any point within the chain-the inventory management module, for instance-means that damage ripples throughout the whole system, leading to disturbances in order processing, shipment scheduling and even customer service. This is undesirable in large distribution centers
8, where literally every minute of system downtime translates into massive financial losses, not to mention fissures in hard-earned customer trust.

Additionally, traditional WMS architectures face difficulties in integration with external systems. As supply chains become more interconnected, the need for seamless data exchange between WMS and other systems such as transportation management systems (TMS)
9, enterprise resource planning (ERP) systems10, and supplier management systems11 has become critical. However, traditional WMS architectures often rely on point-to-point integrations that are not only complex to manage but also prone to errors and data inconsistencies.

Another challenge faced is the adaptability of the system to changes in regulations and compliance requirements. With international boundaries on supply chains, the WMS system needs to comply with varying customs regulations, trade compliance standards and residency laws for data. Traditional systems are rigid and require extensive development and branches in code to include these changes; hence, they are expensive and time-consuming to update.
5. Introduction to Microservices and Their Benefits in Supply Chain and Logistics Management
Microservices architecture12 has emerged as a solution to many of the challenges faced by traditional WMS implementations. Unlike monolithic architectures where all functions are housed within a single application, microservices architecture decomposes the application into a set of loosely coupled and independent services. Each microservice represents a specific business capability, such as inventory control, shipment processing or compliance documentation generation.

One of the prime advantages of microservices within logistics management is scalability. Due to the fact that microservices are independent, scaling up or down will have no impact on other services operating in conjunction. For example, the independent order processing microservice can be scaled upwards during high-volume periods and scaled down during off-peak times without having any effect on any other services running, such as inventory management.

Microservices also support fault tolerance and resilience. For example, in a WMS based on microservices, if one service fails, the damage is confined so that the general system can continue running. This will be very useful in supply chain logistics, where disruptions translate to very expensive delays.

Other beneficial features of microservices include ease of integration and interoperability. Because the microservices communicate through lightweight protocols, such as REST
13 or gRPC14, the integration with some external systems, APIs, or third-party applications becomes easier. This is effective data interchanging between WMS and other supply chain management systems, which includes the ERP, TMS and CRM system.

Another benefit derived from microservices is the speed of development and deployment. Because each service can be independently developed, tested and deployed without affecting the operation of other services within the system, teams can implement new features or updates. This level of agility is very important in logistics management since the capability of responding promptly to changes in market conditions or regulatory requirements may mean the difference to competitive advantage.
6. Introduction
The literature review shows the limitations and inefficiencies of traditional WMS architectures, with a particular emphasis on their inability to easily scale, show resilience, and be flexible in their integration with other systems. Such shortcomings become evident in global supply chains where these extensive variances in regulatory requirements, complex logistic operations, and high transaction volumes call for a more modular and adaptive approach. As already pointed out in the review, traditional WMS solutions are monolithic, thus usually suffer from performance bottlenecks and even system-wide failures in case of scaling up or adapting to changing business needs.

This paper proposes the development of a microservices-based architecture to overcome the shortcomings and gaps identified. The proposed architecture uses the advantages of the microservice architecture to independently scale services, better fault isolation, and more seamless integration with external systems, hence addressing the primary challenges discussed. For example, while the related literature review mentions that traditional systems falter during peak demand periods, the architecture described in this paper allows for the horizontal scaling of certain services-such as order processing or shipment scheduling-without affecting the other modules. The proposed architecture avoids the problem of system-wide failures that arise due to the monolithic nature of legacy systems, since each microservice fails independently yet recovers independently, keeping the overall system stable.

More importantly, practical benefits from transitioning towards a microservices-based WMS architecture through case studies from Amazon, Alibaba and Walmart prove its feasibility in the real world. Most of these organizations have already implemented microservices for their WMS, but in-depth documentation or frameworks that guide the architecture and interactions of microservices within a WMS are still lacking for mid-sized and growing enterprises. This paper tries to fill that gap by providing concrete architecture design, flow diagrams and explanation of the role and responsibility of each microservice in the system. It will provide a basic framework for those organizations willing to adopt a microservices-based WMS but won't have any structured reference model.

Moreover, even though the related literature review identifies microservices as the solution, it does not explore how the architecture can be optimized for specific supply chain operations, such as managing international shipments or compliance requirements related to a certain region. The paper fills this void by offering, in detail, a step-by-step breakdown of the manner in which microservices can be grouped and deployed in distinct environments, such as the Origin Country and Destination Country. By grouping services such as Inbound Services and Outbound Services according to functional boundaries - the architecture ensures a match between the system, with its varied regulatory requirements, and streamlines international logistics operations.
7. Process Overview
The proposed architecture comprises certain major functional domains which are Core Services, Inbound Services, and Outbound Services interacting with external entities like International Carriers. Core Services shall maintain core operations: Inventory Control, Resource Management and Task Management. This ensures that the inventory tracking is efficiently done, the resources are well allocated, and the coordination of the tasks within the warehouse is also carried out efficiently. The Inbound Services shall be responsible for receipt, inspection, storage and management of incoming shipments, value-added services on inbound, compliance regulations and sorting and consolidation. Outbound Services shall take care of domestic and international logistics, which includes Export Customs, Domestic Transport and Deconsolidation, preparing shipments for delivery based on order priority and destination. With its modular microservices interacting clearly with entities outside itself, this architecture offers increased flexibility, scalability and resilience for the system and would be very suitable for complex global supply chain operations.



Below is the detailed flow diagram that represents the interactions between the microservices in the proposed WMS architecture. The diagram highlights operations within the Origin Country and Destination Country environments and illustrates how the product flows through each service, starting from product creation to final delivery.
8. Origin Country Operations


Seller creates a product, which triggers Core Services to update inventory, allocate resources and create tasks.

Inbound Services processes the incoming shipment, applies necessary labels and quality checks and consolidates items based on destination and category.

The product is handed over to Outbound Services for domestic transport and export customs clearance, before being transferred to International Carriers.

Destination Country Operations

 



International Carriers deliver the shipment to the destination country, where Inbound Services handle trade compliance checks and import customs clearance.

After customs clearance, Outbound Services manage de-consolidation, repackaging and value-added services before arranging domestic transport for final delivery to the customer.
9. Service Overview
Core Services
The Core Services form the backbone of the warehouse management system, providing a centralized functionality set to manage inventory, allocate resources and orchestrate tasks. This is an important component in view of seamless integrations among other microservices and assurance that all the fundamental operations within the warehouse are performed in due form. A more detailed breakdown of its sub-services and responsibilities follows below.

Inventory Control Service
This service shall maintain the integrity of product inventory in the warehouses and ensure that the product inventories are updated for every transaction regarding new arrivals, outbound shipments and internal transfers in real time. It integrates into a variety of other systems, including point-of-sale18, transportation management and ERP systems for a unified view of inventory.

Real-time Inventory Updates
: The service is always on watch, constantly updating the inventory count in real time to make the precise representations of stock available. It keeps track of how much of the product is available, reserved and in transit to maintain an updated view of warehouse stock status.

Stock Allocation:
During an inbound shipment, the service selects the most appropriate storage location according to the product characteristics, stock available in the warehouse and other storage constraints. Similarly, for outbound operations, it assigns stock based on shipment priorities, customer orders and SLAs.

Inventory Reconciliation:
It periodically reconciles the actual stock count against system records to identify variance. This can enable alerts identifying lost or damaged items and trigger reordering of those items that have reached a specified low-stock threshold.

Inventory Across Warehouses:
Inventory is kept consolidated across all warehouses and cross-warehouse fulfillment is allowed. It can thereby allow the system to route orders to the most appropriate location based on the availability of the stock with minimal delay and at reduced costs.
Integration with external systems includes sharing data on inventory with other external systems such as ERP, CRM, and vendor management systems for the smooth flow of information that enables the concerned stakeholders to maintain accurate stock positions and make better plans for procurement and sales forecasting.

Resource Management Service
The Resource Management Service will manage the mechanism of allocation and scheduling in the workforce, warehouse space, containers, bins, printers and other warehouse equipment. It also optimizes resource utilization in line with real-time demand and workload for seamless operations.

Workforce Allocation: This service will know who from the warehouse personnel is available and their skill sets, which will create an appropriate workforce allocation. Skilled forklift operators can be utilized where heavy lifting is required, while the other skilled quality control personnel can be utilized for the sensitive shipment inspections.

Equipment Management: This service will set up the machinery/vehicle availability inside the warehouse through scheduled maintenance activities, thereby following up with equipment status to avoid failures during peak operational periods.

Space optimization: It monitors continuously the consumption of space in the warehouse and assigns storage areas dynamically depending on the shipments that arrive or leave. The service wants to make the best use of the shipment depending on the product's size and weight, among other elements, and conditions of storage like temperature-controlled areas.

Demand Forecasting and Capacity Planning: The service forecasts future demands of resources using historical data and predictive models. This allows the warehouse to get ahead of time with the number of staff and the availability of equipment, suggesting the addition of temporary workforces in high workload seasons or storage area reconfigurations.

Data integration with HR and Asset management systems will ensure that resource allocation data is shared for automatic scheduling, payroll, and asset lifecycle maintenance.

Task Management Service
This is the orchestration engine that governs all activities across a warehouse in terms of creating, assigning, tracking and completing tasks. Such a service ensures that each step in the inbound and outbound processes is executed with efficiency and effectiveness.

Task creation and assignment: The Task Management Service automatically creates the respective tasks in case of an event, such as an arrival of a shipment or an order, and assigns them to personnel or equipment according to predefined rules and actual availability.

Scheduling and Prioritization: The service prioritizes tasks based on their urgency, the order value, and the SLAs. For example, it prioritizes perishable goods ahead of non-perishable items and gives express shipments priority over standard deliveries.

Workflow Coordination: It coordinates respective tasks that are spread across services to ensure that any dependencies that may exist between different operations are respected. For example, the service will not schedule a shipment for dispatch until it has passed through quality inspection.

Task monitoring and tracking: It updates the progress of each task in real time for supervisors and managers to track productivity while finding out the bottlenecks. Any delay or problems are flagged for timely resolution.

Performance metrics/reporting: It captures metrics such as task completion time, resource utilization, and error rates that can be used for further action. It generates reports and dashboards that help identify inefficiencies and enable process improvement.

Inbound Services
The Inbound Services are a group of services that are responsible for handling all operations related to receiving, processing, and storing incoming shipments at the warehouse. It ensures that products are correctly documented, inspected, and stored in appropriate locations to maintain operational efficiency and product quality.

Shipment Processing Service
The shipment processing service acts as the entry point with respect to incoming shipments, and it assumes responsibility for the receipt, verification, and documentation of all incoming shipments.

Receipt of shipment and documentation: On arrival to the warehouse, the shipment details are checked against the purchase orders for carrier information, shipping method and expected items. These are documented and sent back to the supplier for resolution in case of discrepancy.

Quality Inspection and Damage Control: The service embarks on quality inspections according to the product type and history with the supplier. For instance, fragile items or high-value items will have more checks to ensure that there is no damage. Goods which are damaged are set aside for return, reorder or special processing.

Product Labeling and Identification: The products are labeled with barcodes or RFID tags including information like SKU, batch number and expiration date in the case of perishable items. The label identifications are used for tracking and identification at the time of storage and retrieval.

Carrier System Integration: Confirmation of delivery, delay tracking, or Advance Shipment Notifications (ASN) can be sent to carriers, alerting the warehouse of a shipment coming so planning can take place.

Inbound Value-Added Services (VAS)
Inbound value-added services involve some of the other activities that add value to a product before storage or dispatch: labeling, re-packaging, and customization of products.

Product Relabelling and Repackaging: Products may be required to have relabelling or repackaging done in compliance with regional regulations or as per customer needs. A shipment of a bulk lot might be divided into smaller lots and labels translated in different languages.

Value Addition and Customization: The activity can also concentrate on promotional stickers additions, kitting, bundling, and product personalization16. These value-added activities not only enhance the product's value but also customize it for market-specific requirements.

Compliance and Documentation: Ensures the conformance of products to regulatory requirements in addition to providing the necessary documentation, including SDSs or Certificates of Compliance17.

Sorting and Consolidation Service
The Sorting and Consolidation Service sorts the incoming product by category, destination or order priority before it optimizes storage and prepares the products for downstream processes.

Categorization and Destination Sorting: The products are categorized based on predefined rules of type, SKU or customer destination. This minimizes retrieval time during outbound sorting.

Consolidation for efficiency: Related products go together to be handled and retrieved with much more ease. For example, items to be shipped out to the same region can be put together to reduce picking and packing time.

Replenishment Planning: If the product is a recurring order, then service plans for its placement in the warehouse to optimize for replenishment.

Outbound Services
The Outbound Services are a group of services that manages all outbound logistics, including preparing products for shipment, handling domestic and international transportation and ensuring that all necessary documentation is in place. These groups of services manage the following responsibilities, since use-cases for outbound activities vary, the following section shows how the outbound services can be leveraged in the WMS.

Logistics Transportation Management: transportation management of the movement of goods within the country of origin either to local customers or to ports for international transport.

Shipment Scheduling and Dispatch: Contacts carriers for the booking of collections and dispatches products on site, ensuring that the product is ready. It takes into consideration variables such as the volume of shipment, mode of transport and destination.

Routing Optimisation and Carrier Selection: Recommends the most viable routings and selects carriers based on cost, reliability and delivery timeframes. For instance, urgent shipments can be scheduled with express carriers.

Tracking Shipments and Monitoring: This service deals with real-time shipment tracking and informs customers and all other parties concerned about shipment status. Delays or problems that could arise are brought out for proactive resolution.

Preparation and Verification of Documents: It prepares the export documents needed such as commercial invoices, packing lists, and certificates of origin. Additionally, it verifies whether these documents can meet the regulatory requirements of the importing country.

Compliance and Tariff Calculation: Product classification review, calculation of duties, and taxes applicable. Ensuring all shipments are in compliance with trade regulations and perform all value-added pre-clearance activity.

Deconsolidation: This service is responsible for breaking down consolidated shipments at the destination and applying any value-added services prior to product delivery to the customer. This may involve the splitting of large shipments to smaller-size consignments as needed. This, in particular, is the case for bulk shipments that need to be divided among a number of consuming customers or locations.

Final Value-Added Services: Such activities may include repackaging, adding local language labels, or promotional material specific to the market destination.

Integration of Microservices in the Overall Flow
Each microservice works in tandem with others through a well-defined event-driven architecture19. Events such as "Shipment Arrived" or "Product Created" trigger a series of activities and notifications between services. For example, when InboundServices receives a shipment, it triggers Core Services to update inventory and allocate storage resources. Similarly, when OutboundServices schedules a dispatch, it notifies International Carriers for handover, ensuring a smooth transition across all stages of the supply chain.
10. Case Studies: WMS Transformations in Large Organizations
The microservices-based WMS architecture for scalability, resilience, and integration has been adopted by several large organizations in their supply chain operations. The following are some case studies of how such a change to microservices architecture has affected them:

Amazon
Globally, Amazon is renowned for having one of the most automated and scalable warehouse management systems. It chose a microservices architecture for its WMS to support the huge volume of orders processed daily across its global distribution network2. This means inventory management, order processing and shipment scheduling are independently scaled services within an Amazon WMS based on microservices architecture. This loose coupling of services further enables Amazon to optimize performance with less disruption during peak demand. The seamless integration of the WMS with the rest of the logistics ecosystem at Amazon allows for real-time order tracking and dynamic route optimization.

Alibaba
Cainiao Network, the logistics arm of Alibaba Group, refactored its WMS into microservices architecture to meet the growth of its e-commerce business. The new architecture enables Cainiao to serve complex logistics operations in China and beyond, guaranteeing timely deliveries during massive sales events down to the annual 11.11 Global Shopping Festival3,4,5. The module-based nature of the microservices architecture allows Alibaba to enable the use of advanced technologies such as AI and ML for predictive analytics. That would mean the company can predict order surges in advance and adapt the logistics operations accordingly.

Walmart
At Walmart, WMS transformation involved the migration away from the traditional SOA-based architecture toward a microservices architecture to meet its supply chain scale. The new system will also further enhance multi-distribution center inventory management, route optimization of shipments, and integrating with various supplier systems. Walmart's WMS transformation also placed much-needed emphasis on system resiliency to ensure that each microservice was independently able to recover in the event of failure6,7. This architecture allows Walmart to sustain very high levels of operational efficiencies and reductions in order processing times, thus improving customer satisfaction.
11. Conclusion
This paper proposes a microservices architecture that is tailored for use in warehouse management systems, hence allowing for a modular, scalable, and fault-tolerant approach in the management of complex global supply chains. It allows for efficient handling of high-volume operations due to the fact that it is divided into independent microservices based on functional domains scaled or deployed at any given time in the need-based requirement. Splitting the functional architecture by country of origin and country of destination environment further enhances the architecture's capability to address region-specific logistics and regulatory requirements.

The architecture will be helpful in two ways: first, simplifying the management of large-scale operations and second, improving resilience by addressing how one would recover from failures or disruptions in specific services. Subsequent work can focus on AI-based predictive analytics for optimization of resource allocation and enhanced decision-making processes down the supply chain.
12. References

  1. Lee CKM, et al., Design and Application of Internet of Things-Based Warehouse Management System for Smart Logistics, International Journal of Production Research, doi: 10.1080/00207543.2017.1394592. 2017;56(8):2753-2768
  2. How to Build a Dynamic Supply Chain Platform: A Primer, AWS, Available: https://aws.amazon.com/blogs/industries/how-to-build-a-dynamic-supply-chain-platform-a-primer/
  3. Cainiao Network Technology, Alibaba Group, Available: https://www.alibabagroup.com/en-US/document-1491566211274637312
  4. Alibaba Cloud and Cainiao Network: Digital and Intelligent Upgrade of the Entire Logistics Chain, Alibaba Cloud Community, Available: https://www.alibabacloud.com/blog/alibaba-cloud-and-cainiao-network-digital-and-intelligent-upgrade-of-the-entire-logistics-chain_595414
  5. Cainiao Business Overview, Alibaba Group, Available: https://www.alibabagroup.com/en-US/about-alibaba-businesses-1489025287453671424
  6. Richard Rodger, The Tao of Microservices , Manning, 2017.
  7. M. Vizard, Walmart Embraces Microservices to Get More Agile, Baseline 2015, Available: https://www.baselinemag.com/enterprise-apps/walmart-embraces-microservices-to-get-more-agile/
  8. Distribution Centers, NetSuite, Available: https://www.netsuite.com/portal/resource/articles/inventory-management/distribution-centers.shtml
  9.  A. Lazanas and G. Megalokonomos, Optimizing Alternative Routes Retrieval in an Agent-based Transportation Management System, in 2006 International Conference on Service Systems and Service Management, Troyes, France, 2006, pp. 1525-1530, doi: 10.1109/ICSSSM.2006.320762.
  10. A. Farah, E.-C. Jihane and N. Hmina, "The importance of enterprise resource planning (ERP) in the optimisation of the small and medium enterprise's resources in Morocco," in 2022 8th International Conference on Optimization and Applications (ICOA), Genoa, Italy, 2022, pp. 1-6, doi: 10.1109/ICOA55659.2022.9934158.
  11. Supplier Management, IBM, Available: https://www.ibm.com/topics/supplier-management
  12. Y. Romani, O. Tibermacine and C. Tibermacine, Towards Migrating Legacy Software Systems to Microservice-based Architectures: a Data-Centric Process for Microservice Identification, in 2022 IEEE 19th International Conference on Software Architecture Companion (ICSA-C), Honolulu, HI, USA, 2022, pp. 15-19, doi: 10.1109/ICSA-C54293.2022.00010.
  13. What is RESTful API? AWS, Available: https://aws.amazon.com/what-is/restful-api/
  14.  gRPC Overview, gRPC.io, [Online]. Available: https://grpc.io/
  15.  A. Žák and M. Bobák, "Modular e-Commerce Data Warehouse using Microservices," in 2022 Cybernetics & Informatics (K&I), Visegrád, Hungary, 2022, pp. 1-6, doi: 10.1109/KI55792.2022.9925962.
  16. Value-Added Services, Service Lobby, Available: https://service-lobby.com/en/value-added-services/
  17. Export and Import Compliance, Curtis, Available: https://www.curtis.com/glossary/international-trade/export-and-import-compliance
  18. CJ Debono and D Busuttil, A secure wireless point of sale system, in 2011 IEEE EUROCON - International Conference on Computer as a Tool, Lisbon, Portugal, 2011, pp. 1-4, doi: 10.1109/EUROCON.2011.5929195.
  19.  A Rahmatulloh, F Nugraha, R Gunawan and I Darmawan, Event-Driven Architecture to Improve Performance and Scalability in Microservices-Based Systems, in 2022 International Conference Advancement in Data Science, E-learning and Information Systems (ICADEIS), Bandung, Indonesia, 2022, pp. 01-06, doi: 10.1109/ICADEIS56544.2022.10037390.