6 Factors for Real-Time Integration…Buzzword or Real Need?
Real-time processing implies that processing of data happens upon the completion of an event (request for data for example) and is immediately handed off to the next system that processes immediately and so on. Here, I’m mainly referring to the flow of information through the supply chain and the speed at which that information travels, which helps orchestrate the flow of materials and resources. Access to information in real-time provides business intelligence that allows decisions to be made about any changes that may be needed for the orchestration.
An example always helps so let’s look at something I come across at least a few times a month:
You have an E-Commerce store and you want your customers to have immediate access to your inventory levels so you can provide accurate stock levels and shipping dates.
Real-time communications between your E-Commerce store and ERP or inventory system can inform the customer that the last pair of sunglasses was sold 2 minutes ago and is now backordered (or shipping in 3 weeks when more are in stock).
What makes this scenario tricky is that your E-Commerce inventory levels are also affected by external functions such as telephone, email or fax orders keyed directly into the ERP system, re-allocation of inventory to another warehouse and discovery of damaged goods to name a few.
How much transparency do you want your customers to have on stock levels when placing an order and how will this be presented on your E-Commerce store?
What does the customer see if your point-to-point real-time inventory query system is down?
What if your warehouse is outsourced and won’t allow you to connect your E-Commerce store to their systems directly?
How easily will you be able to add, remove or change the way the information flows once the system is running?
The 6 Factors
1. Speed – The biggest benefit and main purpose for real-time integration is speed. If an order is placed on your E-Commerce system and that data is made available to the warehouse to pick, pack and ship within a few minutes, you can greatly reduce your turn-around time and process orders closer to the cutoff time for shipping.
2. Transparency – If you can view the key metrics in your business you need to make informed decisions at the very moment things are happening, you can make better decisions earlier, such as forecasting potential inventory shortages so you can contact your suppliers for more products sooner.
3. Cost – Depending on scope, the cost of the infrastructure required need to support a heterogeneous real-time network would probably be prohibitive in most cases. Also consider future costs of updating or changing any system in the process since connections between real-time systems today are more likely to be ‘hard-coded’.
4. Complexity – The objectives for a real-time project should be clearly defined and all systems need to be identified and taken into account. This includes the number and types of systems through which the data travels (both hardware and software) and will likely impact everyone in your organization.
5. Risk – Point-to-point integration means that ANY changes to any system in the process would need to be managed very carefully. Without a processing queue, any failure or hiccup in any system in the chain could result in data loss.
6. Flexibility – The architecture of a real-time system is not usually as flexible to changes as batch-based systems.
I can’t help but wonder if the desire for real-time integration processing is a bit of a buzz-term, at least partly driven by our almost instant and universal access to information. When you’re the size of companies like Wal-Mart and Ikea, real-time processing is indeed very relevant, but for many small and medium-sized businesses, substantial gains can usually be made by looking at efficiencies in your existing business process.
What are your thoughts on where we are heading with real-time processing in the supply chain? Leave me a comment.