Data quality in telecoms network inventory management

23 July 2020

Data quality in telecoms network inventory management

Data quality has a huge impact on service performance and is fundamental to effective inventory management. But, maintaining effective data quality can be a challenge. Find out more and how to solve this problem with CROSS.

Problems with data quality undermine service delivery

Inventory management is a key function across many sectors, but it’s particularly important in telecoms. To understand why, let’s explore a problem that has been analyzed in depth, here.

The example described highlights an issue that was investigated manually. Let’s imagine a warehouse, in which items are stored prior to dispatch to a required location. An order is received which requires a particular item as part of its fulfilment. According to the inventory system used in the warehouse, there are 10 of these items in stock. Accordingly, a picker goes to the shelf and discovers that, in fact, there are only two.

This has several impacts. First, the capacity to meet future orders is impaired. While this particular order can be met, only one future order can be delivered. If stock is managed at a rate to keep pace with orders, delays will lead to a failure to meet customer expectations or delivery performance requirements. The situation will escalate as the orders mount up, because the inventory level is lower than the appropriate threshold. Problems will cascade.

Second, we need to think about why the inventory data is incorrect. As the author notes “they [the staff] consider two possibilities: (1) the remaining eight units are around somewhere, they have just been placed in the wrong location, and (2) the system quantity has gotten off track and is now incorrect.”

A fishbone diagram is used to illustrate the case and the analysis of possible causes.

Figure 1 – Fishbone analysis of incorrect inventory data

fishbone analysis

He concludes: “investigation reveals that the items are nowhere to be found, and attention is turned to the second possibility”.  Next, attention is focused on the reasons why the stock levels may have got out of kilter, for which there can be several reasons, as the diagram reveals.

What does this mean in practice?

We don’t need to explore all of these here. Two points should be clear. First, that a simple error has an immediate negative impact on the service delivery performance of the warehouse owner, because the problem has to be resolved, which creates costs. Second, while this delivery can be completed, a bigger problem is coming up, at a rate determined by the flow of orders.

Our example is a simple warehouse and relates to a single item. We do not know whether the item is part of a series of items that are required to complete the order; it might be, it might not be. It could be that this is a critical element, without which the order cannot be completed, or it may simply be something that can be deferred, allowing partial completion. It doesn’t really matter – any or all of the above could be true. We need to manage inventory so that we avoid all of these problems.

inventory management

Telecoms inventory for service delivery

Telecoms networks need something like a warehouse. As we’ve noted before, a service consists of a series of logical elements. All telecoms networks are composed of logical building blocks that are connected in the correct order to deliver a service. A service can be as simple as a fiber connection to a specific address, or it could be a subscription to a service that combines different elements (voice, data, messaging) with the units that are required for their delivery (a subsidized device, with a contract).

Each time a service is ordered, someone needs to visit the warehouse to retrieve and allocate the required components. Each time the network is extended, the same needs to happen. The Bill of Materials (BoM) may be different in each case, or it may have common components. But, in each case, you need to know what you have in the warehouse so that you can determine if an order can be delivered.

This is a relatively straightforward process in a small organization, but it still requires the right tools to ensure effective inventory management. If we scale this up to the size of a typical telecoms network – with, for example, several millions of customers – the complexity of managing an inventory that spans all of the elements used to extend and enhance the network and to deliver customer-facing services should be apparent. We need a sophisticated warehouse.

Enhancing data quality for telecoms inventory management

And, the importance of data quality should also be apparent. If you only have two units instead of 10, something has gone wrong, somewhere. The data needs to be corrected. If the details were kept on a notepad (astonishingly, many operators still retain paper or spreadsheet-based records), it’s easy to imagine how this might happen.

But, while rectifying this data fault is one thing (you could just manually change the listed number to reflect the real situation), there’s another dimension. Can the error be corrected automatically? That is, can the required stock level be met when the fault is discovered by automatically ordering the missing eight, so that equilibrium (normal stocking levels) be restored?

Data quality can thus be seen to be essential to any effective inventory system. A single incorrect entry can be a problem, but thousands of incorrect entries could be catastrophic for the success of any operator. Data quality needs to be maintained. For many operators, this is a huge challenge because they have a legacy of poor-quality data – while the problems this can lead to are known, the challenge of cleansing this data is seen as insurmountable.

The telecoms data warehouse

Fortunately, there is an answer. CROSS is a complete network inventory solution that allows data to be imported from any source. However, it also allows data to be tagged as it is imported, so that it is assigned an indicator of the presumed quality. Data that is known to be accurate can be safely processed, while data that is suspected of some flaw can also be examined independently. In other words, it’s possible to segregate data so that the precise number of widgets on the shelf can be determined – before they are required.

And, because automated and manual data entry are both possible, it’s easy to ensure that any operative can update the records. Crucially, CROSS doesn’t just record physical inventory assets, it also holds records of logical, virtual and service assets. This means that a BoM can be assembled that encompasses all of the elements required for network extensions and service deliveries, quickly and easily, with any known issue clearly identified.

This also enables root-cause analysis, as we discussed recently. CROSS provides a comprehensive solution that enables data quality to be identified, enhanced and maintained, making it available to other operational systems that depend on inventory information to build orders and deliver services to customers. Think of it as a virtual warehouse for your telecoms network.