Menu

Professional    Learning    Academy

From the experts in Telecommunication, Cyber Security, Big Data and IoT

icon icon

Learn Cellular Wireless communication technology, its architecture and management

icon icon

 

Learn construct of Information security management system in a simple way

icon icon Learn application of Big data analytics in Internet of things domain

Big Data Analytics in Supply Chain

Supply chain is very funny terminology. For one company, what distribution of product/services is, for other company it is the supply chain.  This chain is ever growing with business model innovation. The supply chain refers to the acquisition of raw material, operational resources like infrastructures, equipments, spares etc and services required for a business in order to produce the finished product and/or services. Although loosely defined, but high level, this is what the supply chain is all about. In a B2B2B2B2B….2B2C, the preceding B for any B is the supplier. Though, it is simplest but complex relationship can co-exist.  

The question that is going to be put before us is ,How big data analytics helps supply chain of an organization, also known as SCM.The Bottom line is, Big data is providing supplier networks with greater data accuracy, clarity, and insights, leading to more contextual intelligence shared across supply chains.

Forward-thinking manufacturers are orchestrating 80% or more of their supplier network activity outside their four walls, using big data and cloud-based technologies to get beyond the constraints of legacy Enterprise Resource Planning (ERP) and Supply Chain Management (SCM) systems. For manufacturers, whose business models are based on rapid product lifecycles and speed, legacy ERP systems are a bottleneck.  Designed for delivering order, shipment and transactional data, these systems aren’t capable of scaling to meet the challenges supply chains face today. Choosing to compete on accuracy, speed and quality forces supplier networks to get to a level of contextual intelligence not possible with legacy ERP and SCM systems.

 

Procurement deals with the relationships at the upstream supply chain. Data complexities on this side might arise from globalised purchasing strategies with thousands of transactions. In this lever, a strong connection with internal finance reporting led to adopt measures on spend visibility data, to achieve granular levels on aggregated procurement patterns. Nevertheless, data on external expenditure, which can be more than 50% of a company’s cost, are “often backward looking, often inconsistently categorised and not integrated with internal costs”. A subgroup of data that is still to be fully integrated and appears in the taxonomy as semi-structured are the business documents (purchase orders, shipping notices, invoices) sent through the EDI. The procurement needs to activate  the data sources not only for spending data management process, but also for the entire procurement function.

Warehouse management (particularly inventory management) has been radically changed by modern identification systems after successful introduction of RFID. The largest clusters of data are related to an automated sensing capability, especially as the Internet of Things and extended sensors, connectivity and intelligence to material handling and packaging systems applications evolved. Position sensors for on-shelf availability share space with traditionally SKU levels and BOMs.

Transportation analysis applying Operational Research models has been widely used for location, network design or vehicle routing using origin and destination (OND), logistics network topology or transportation costs as static data. New alternatives to manage and coordinate in real time using operational data rely on mobile and direct sensing over shipments that are integrated into in-transit inventory, estimated lead times based on traffic conditions, weather variables, real time marginal cost for different channels, intelligent transportation systems or crowd-based delivery networks among sources of Big Data. A detailed analysis of the 3 Vs in transportation data revealed to be the facilitator for proportionally higher speeds in data transition.

Based on the above criteria four such area of application with input data and outcome in supply chain domain has been described below:

 

SCM lever Functional problem Type of data BDA proposed solution BDA techniques

  1. Marketing Sentiment analysis of demand- Data on New trends, blog, news feeds, rating and reputation from 3rd parth, web logs , loyalty programs, call centres records, customer surveys can be analysed to identify key product demands using the well known NLP, statistical models and supervise machine learning methods like Logistic regression, Random forest, k-NN, Naïve Bayes etc.
  2. Procurement – Data on supplier negotiation, supplier relationship management,  Supplier current capacity & top customers, supplier financial performance, data regarding previous transactions of the supplier with other third parties in similar characteristics (delivery locations, lead times) can be analysed to derived to device  the  performance requirements for procurement contracts (SLA or other quality measures).
  3. Warehouse Operations Warranty Analytics --  Data on Internet of things sensing, user demographics, historical asset usage data can be aggregated from multiple sensing sources on real time clubbed with reports on monitored assets can provide the vital clue for organization to check out the warranty compliance of spares as well segmentation of inventory based on promised and actual warranty compliance. The space utilization and occupancy provides the statistics on vital inventory holding parameters.
  4.  Transportation -  Data on Real time route optimisation, Traffic density, weather conditions, transport systems constraints, intelligent transport systems, GPS-enabled Big Data telematics can be analysed in order to address time variability for deliveries in predefined networks, model the delivery network and update it with current position of delivery units. Also, new requirements for delivery keeps on adding into the system, so taking into account all network availability factors, from each delivery unit a spatial regression analytics can predict the  time and cost of serving a delivery to other point of the network

Needless to say, Big data analytics is enabling complex supplier networks to focus on knowledge sharing and collaboration as the value-add over just completing transactions. Organization has increasingly started using big data analytics and  statistical process control to increase supplier quality from supplier audit to inbound inspection and final assembly.

In summary, big data is having an impact on organizations’ reaction time to supply chain issues, increased supply chain efficiency, and greater integration across the supply chain.

Go Back

Comment

Blog Search

Comments


Blog Archive