Empirical studies in supply chain management

Empirical research in manufacturing and service operations management is an emerging and promising area of research with potential to contribute significantly to our understanding of companies\u27 operational practices. In the first part of the dissertation, we describe a study of the consequences of capacity allocation mechanisms at a large semiconductor manufacturer. We use a panel data set with 300,000 observations spanning five years of forecasting and order transactions between a supplier and all of the customers at the supplier\u27s manufacturing facility. We find empirical evidence that a supplier\u27s capacity allocation policy distorts customers\u27 demand forecasts and orders. These distortions are of two types, (1) customer forecasts overstating the underlying demand\u27s actual volatility, called churn and (2) customers\u27 forecast updates exhibiting spurious correlation amongst themselves, called contagion. The findings have implications for contracting, capacity planning and CPFR initiatives, in supply chains. In the second part, we study retail store execution at two anonymous retailers, we call Alpha and Beta. Store execution encompasses all activities within a retail store that impact the store\u27s profitability and customers\u27 satisfaction. We identify action steps that are likely to improve in-store execution measured using variables such as sales, customer satisfaction and in-stock rate. For retailer Alpha, we develop a three-stage econometric model to analyze the marginal effects of various execution levers on sales, customer satisfaction and customer perceived in-stock. Our model explains approximately 75%, 97% and 71% respectively, of the residual variation in these variables after accounting for seasonality, time trends, autocorrelation and forward looking behavior. We find that sales is primarily driven by actual in-stock, overall satisfaction and payroll level. For retailer Beta, we show that customers\u27 average basket value, not sales, is an objective measure of store performance. We explain 53.3% of the variation in customers\u27 average basket value at different stores. We identify customer demographics, temporal mismatch between planned payroll and customer traffic and the temporal mismatch between actual payroll and planned payroll as significant drivers of average customer basket value