View all articles

How we manage supply chain complexity

By Stefan Gerber

Supply chains are the backbone of countless industries, ensuring the smooth flow of goods and services from manufacturers to consumers. In today's rapidly evolving business landscape, organisations are constantly seeking innovative ways to optimise their supply chains and gain a competitive edge.

In this post, we will explore the capabilities of graph data solutions and how they can provide benefits to the way organisations view and operate their supply chains.

Let’s start with an example. A company manufactures air conditioners at two plants in different countries; parts are sourced from multiple suppliers and shipped to customers across the globe. To manage the data, different entities may be separated as follows:

A simplified entity relationship diagram (erd) depicting different business units within a simplified air conditioner manufacturer's supply chain.

This is a simple example of an entity-relationship diagram (ERD) for a small number of entities within this supply chain. A more realistic one would look something like this:

A large enterprise entity relationship diagram (erd) with dozens of connections.

Another way to view these diagrams would be to think of all of the lines connecting the various tables as an added dependency. If you, for example, needed to get from the customer's table to the distribution centre table, you would need to first join a customer to a shipment ID (following the orange line) within the logistics/transport table and a shipment ID to a distribution centre (following the black line). In real-world use cases, the number of dependencies could quickly grow to hundreds or thousands.

Consider a scenario where customers start complaining about air-conditioners dying within a few weeks of receiving them. This could quickly become a major issue for the business, resulting in a loss of reputation and brand credibility. To prevent this, we need to investigate and identify the source of the failures as soon as possible. We must find the defective part responsible for the issue and identify the supplier, manufacturer, distribution centre, and potentially affected customers within the supply chain.

It is better to be proactive in our business practices than to react when dissatisfied customers complain.

So, from our example, we would need to walk through our table of relationships to find out where these failing parts originated from and, more importantly, where they’re located in our supply chain at this very moment. Solving this problem could be very difficult, depending on the complexity of the relationships and data access within the organisation. And since the records from our example are stored in tables, they would need to be scanned from top to bottom for every customer, order, product, delivery etc. This can be a computationally expensive exercise (when dealing with millions or billions of records) and, depending on the size of the operation, costly in terms of resources (financially and labour wise).

Supply chains are connected. Supply chains are graphs #

By taking a step back, we can visualise our dependency table differently. visual graph depicting depicting the entities and relationships within the manufacturers supply chain.

By defining a graph model of our supply chain example, we’ve created a consolidated view of the business entities. It becomes much simpler to navigate and understand by defining entities and dependencies in an (Entity) connected by a [Relationship] to another (Entity) form, as shown above, than it does by connecting tables and scanning through records.

Each entity within our model would also contain properties describing for example it's unique identifier, shipment time, product type, part number etc.

Let’s dig into some examples of how this can be beneficial to large organisations.

Important questions to ask within our supply chain would be those able to give us information on product trends and demand throughout the company’s distribution network. We could do this analysis by combining orders, product and distribution centre data.

visual graph depicting depicting a subset of the entities and relationships within the manufacturers supply chain, specifically for the products, orders and distribution centres.

In this example, we have zoomed in on two products with four orders between them. From the relationships, we can also see that they’re stored at three distribution centres within our supply chain, one of which they share. Depending on the size of the distribution network, there could be hundreds of similarly shaped product, order and distribution centre combinations.

Using these entity's and their properties and connections, we can find out the following:

  • The current best sellers within a region
  • Common properties of best sellers within regions
  • Low inventory risks at distribution centres within best-seller regions
  • How and when products from low-performance regions can be shipped to high-performing regions to reduce low inventory risk
  • Order fulfilment status, and which distribution centres are performing well, or not, with regard to order fulfilment

More advanced examples would be to use the capabilities of machine learning and graph data intelligence to perform demand sensing. These algorithms would be able to take into account market data and sales to drive insights and determine risk throughout a supply chain and include:

Supply Management #

The ability to forecast inventory levels using historical purchase behaviour and current demand to quickly determine alternate suppliers or distribution centres when local options lack the necessary capacity.

Dynamic Risk Alerting #

The ability to connect to and monitor alternate data sources (market data, news, weather, traffic, etc.) and apply risk scores to the current supply chain network.

Business Planning #

The ability to perform scenario and demand planning using a digital data-driven twin of your supply chain allowing your organisation to balance demand, supply, inventory, and financial targets.

High-Value Analysis #

Powerful and flexible queries allow your organisation to analyse the interactions between specific entities within your supply chain. For example, you can identify trends between products, customers and regions within minutes.

This new capability enables organisations to identify risks within their system to plan and prepare for worst-case scenarios. Large organisations fully committed to vertical integration can also perform cross-chain analysis to improve service delivery and customer satisfaction.

Conclusion #

Managing supply chains are a complex problem due to the sheer volume of dependencies and cross-connected entities. To operate effectively, business leaders and operators need to understand and manage risk within their systems.

At Rockup, we believe in the power of relationships. If your company can benefit from a deeper understanding of the data that drives your success, contact us to schedule a call.

Subscribe to Rockup’s Insights

We are committed to keeping your data safe.


Too much email?