The innovation is a new method for forecasting the demand for spare and service parts whose demand is intermittent or slow moving. Intermittent demand consists mostly of zero values, with nonzero demands of random sizes mixed in at random times, making it extremely difficult to forecast. The innovation will allow parts providers to operate in the "sweet spot" that balances the costs of keeping unused parts on the shelf against the costs created by not having parts available when needed. The research will develop the new forecasting algorithm, embed it in a prototype software product, and document its greater accuracy compared to conventional methods. The evaluation of accuracy will be based on extensive computational experiments using both synthetic data (to discover which data features are critical) and a library of over 100,000 real-world demand histories provided by existing customers.
The broader/commercial impact will be reduction of a multi-billion dollar drag on the US economy: mismanagement of parts inventories. Parts inventories are the second largest item on the balance sheets of many companies, and spending on them amounts to roughly 8% of the US Gross Domestic Product (GDP). Improved parts forecasting can lead to increases in parts availability by over 10%, and simultaneous reductions of over 15% in inventory costs. These improvements will benefit not only the vendors of parts but their customers, whose supply chains will become more reliable and whose operations will have reduced down time.
Normal 0 false false false EN-US X-NONE X-NONE Project Outcome Report. The service parts market in the US alone is valued at over $1.6 trillion. Unfortunately, this part of the economy has not enjoyed a level of innovation commensurate with its importance, with the result that mismanagement of inventories of spare and service parts creates a multi-billion dollar drag on the US economy. Smart Software’s Phase I SBIR research focused on a key cause of both civilian and military inventory mismanagement: inaccurate forecasts of the majority of parts whose demand is "intermittent" or sporadic. Intermittent demand is highly sporadic and seemingly random, characterized by frequent periods of no demand interspersed with random spikes of highly varying demand, and most common in spare parts inventories. This type of demand is exceptionally difficult to forecast and is generally ignored by demand forecasting software systems. The innovation explored through this research began with the realization that we can achieve greater accuracy in forecasting intermittent demand by moving beyond the conventional one-item-at-a-time approach and adopting a revolutionary multidimensional, multiple-item perspective. This new methodology is based on identifying and exploiting clusters of items whose demand tends to move together, then using the clusters’ statistical interrelationships to forecast each inventory item more accurately. This approach is patent-pending. Phase I concluded successfully with the following major accomplishments: Accomplishment 1: The creation and preliminary evaluation of a working prototype of a new cluster-based algorithm for spare parts forecasting based on the concept of a "personal cluster" for each part derived from co-movements of part demands. Accomplishment 2: The creation of a working prototype of a second new algorithm based on hierarchical clustering of all the parts, again based on co-movements of demand. The research concluded with the recommendation that this work continue, with the objective of developing commercial products that implement these new forecasting methods. A Phase II SBIR application has been submitted, together with letters of support from commercial, semi-public (transit agencies), and military (US Air Force) organizations. The new methods present the potential to significantly reduce inventories, on the order of 15% or more in the first year of implementation, while maintaining or increasing control over the chance of stocking out. For information about Smart Software please visit our website, www.smartcorp.com. To learn more about this research, please send an email request to info@smartcorp.com
