|Roger D. H. Warburton|
Currently, I am conducting research in the following fields.
|Outsourcing & Competitiveness of Domestic Manufacturing|
In this project I developed a formal Quick Response Manufacturing (QRM) business model and successfully demonstrated that for some products that domestic manufacturing is competitive. Indeed, for certain types of goods and services, a domestic manufacturing asset is the only option for long-term profitability.
In lectures and publications to manufacturers, professional societies, and academic institutions, I relentlessly challenge the obsession with manufacturing everything offshore. U.S. manufacturers have consistently observed that the model is one of the few that provides a cost competitive rationale for their existence and is therefore encouraging. A few companies have adopted the model, employing an optimal mixture of offshore and domestic quick response manufacturing.
Consumers are becoming more demanding and markets are increasingly characterized by short product life cycles. Meanwhile, the lure of low cost outsourcing is often pursued without appreciating that long lead times are incompatible with changing consumer demand, and can result in losses from excess inventory or missed sales.
This conflict is especially severe when the product is technologically innovative, because the consumer demand is frequently uncertain, if not downright erratic. In such circumstances, my model provides a strategy to deal with the mismatch between offshore sourcing and the response to changing demand.
For an overview of the how the process works, see:
Warburton R.D.H. and R. Stratton, 2002. “Questioning the relentless shift to offshore manufacturing” Supply Chain Management 7, 2, 101-108.
To calculate the optimal amount of onshore and offshore manufacturing, see:
Warburton, R.D.H. and R. Stratton, 2005. “The optimal quantity of quick response manufacturing for an onshore and offshore sourcing model” International Journal of Logistics, 8, 2, 1 – 17.
|Exact Solutions to the Supply Chain Equations|
In a somewhat surprising development, I recently found that exact solutions exist to the full set of continuous supply chain equations that describe the inventory and orders for a typical industrial information system. No approximations are required.
The order rate can be tuned with three parameters, and they represent fundamental and familiar management quantities: the amount of exponential smoothing to apply to the demand; the rate at which an inventory deficit is to be recovered to avoid stockouts; and the quantity of work in process (WIP) that should be either on order or in production to meet the current demand.
Working with my colleague, Jonathan Hodgson, Professor of Mathematics and Computer Science at St. Joseph’s University, we have validated the analytical solutions with comparisons to numerical integrations. In an important development, my colleague Steve Disney and I have been able to show that the solutions replicate the answers of established discrete models that are the foundation of traditional supply chain management.
In supply chains the ordering policy is used to set production targets, capacity requirements, inventory levels and customer satisfaction levels. My colleague Erland Nielsen and I have determined ordering policies that can react to changes in demand and return the inventory to its desired level.
|Research Example: The Bullwhip Effect|
The Bullwhip Effect is the well-known problem where a retailer’s orders to manufacturers have a much larger variation than the consumer demand. This distortion in demand is continually amplified as orders progress along the supply chain. This so-called “Bullwhip Effect” has been shown to be a significant problem in wide variety of companies and industries. Interestingly, some standard management policies have been shown to exacerbate the effect.
When a surge in demand occurs the challenge is to optimally replenish the inventory without passing on the bullwhip amplification. Using our new solutions, we can calculate the parameters that return the inventory precisely back to its desired value without either deficits or overshoots. Thus we are able to propose strategies that mitigate the bullwhip effect.
The defintion of SPI drove me crazy for a while. At the end of the project, when you complete all of the work the earned value (EV) is equal to the planned value (PV), so the SPI = EV/PV -> 1. Usually, however, one measures SPI < 1, so how does it get to 1?
I developed a method for including time dependence into Earned Value Management (EVM). The model requires three parameters, which map directly to the fundamental ``triple constraints'' of scope, cost, and schedule: the reject rate of activities, the cost overrun parameter, and the time to repair the rejected activities. Time dependent expressions for the planned value, earned value, and actual cost are derived, along with the CPI and SPI. The model is built on the well-established Putnam-Norden-Rayleigh (PNR) labor rate profile, which is a useful representation for large software projects. I apply the model to a well-known PNR software data set, demonstrating how to estimate the project's final cost, which converges faster to the correct answer with less variability than standard Cost-to-Complete calculations. The model also accurately predicts the required revised labor profile and the new schedule.