Franklin Associates, Ltd. (FAL) is developing a methodology to deal with the issues of uncertainty and data quality in Life Cycle Inventories (LCI). In traditional LCIs, single point estimates of input variables (such as fuel requirements) are used to determine single point estimates for the output variables (such as total energy used or solid waste generated). These point estimates contain no information about the uncertainty of the data, and therefore give a false sense of precision.If LCIs are to become more widely used by decision makers and others, an acceptable method of dealing with uncertainty needs to be developed. This paper discusses the data uncertainty methodology being developed at Franklin Associates, and uses a previously completed case study as a real-world example of its use.The FAL methodology involves the assignment of data quality indicators to the variables used as inputs to our computer models. This allows the determination of a distribution of input values, rather than a single point estimate. Our deterministic model therefore becomes a stochastic model, which means that the output of the model is also a distribution of values, rather than a single point estimate. It is then easier to judge, for example, whether two values for total solid waste are the same or different.