IoT’s surprising impact on revolutionizing inventory management

Sarah Hatfield directs OnProcess Technology’s strategy for products and core service offerings, including the OPTvision platform. She brings more than 15 years of leadership expertise from previous roles in supply chain, product and program management for Comcast, Asurion and ADT.

You know disruptive technologies have reached the tipping point when non-IT pros build business plans around them. This is exactly what’s happening with IoT. Because of its ability to drive wide-ranging, game-changing improvements, IoT is starting to be used across all aspects of business operations. One of the newest, and most impactful, areas is spare parts inventory management, a key aspect of the post-sale supply chain.

Maintaining the right level of spare parts is critical. As you can probably guess, carrying excessive inventory can be prohibitively expensive. But if you have too little, you’ll slow product repairs, hurt customer experience and end up spending more money purchasing new parts for stock replenishment. The problem is, traditional best practices for managing spare parts — using time-series algorithms combined with sales forecasting, seasonality, gut instincts and simple rules of thumb to determine how many parts to stock — are woefully inaccurate because:

  • They’re static, “review-and-stock” endeavors based largely on historical demand data
  • The algorithms don’t account for variables resulting from failed parts in the field

Knowing this, many companies hedge their bets by purposefully overstocking. Others think they’re maintaining the right levels, but unknowingly overstock. In either case, they’re wasting a lot of money.

New IoT-driven inventory planning

The key to accurately stocking parts is knowing which ones are likely to fail and when they’ll need to be replaced. Some businesses have attempted to use IoT data to understand product failure impacts on inventory planning. However, most of the IoT monitoring programs are designed to respond to signal failures. Plus, IoT data collection is often haphazard and emphasizes the few pieces of equipment that are starting to fail, rather than the whole. This makes it impossible to generate a sound baseline for analyzing product performance and predicting failures — which, in turn, makes it impossible to accurately forecast spare parts needs.

The good news is there’s a new inventory planning algorithm that builds IoT-based failure data directly into the equation. Developed at Massachusetts Institute of Technology Center for Transportation and Logistics, it enables businesses to accurately forecast needs. By using this methodology and analyzing historical failure data on the entire installed base, businesses can predict the exact spare parts they’re likely to need, when and in what quantity.

The better news is that it doesn’t take a huge team to capture IoT data because not much data is needed…. Read More