Currently, the Philadelphia Water Department is developing a fast, reliable and largely
automated water quality event-detection system with the end-goal of providing actionable
data to concerned members within the utility. To ensure that true water quality anomalies
can be distinguished from daily and seasonal variations, many factors with the potential to
affect water quality must be assessed simultaneously, including the following: pressure/flow in pipes;
construction activity; customer complaints; online water quality monitoring data; and,
laboratory analysis results.
The first step in establishing this system is ensuring that the various data sources can be
accessed from a centralized location. Currently, the data from most of these sources is
stored in a variety of different databases and is only available within the individual
departments where the data resides. The study approach is to establish an Enterprise database where data from all of these sources will be integrated
and stored centrally to be available for utility-wide access by trusted users. This solution
is simple, cost-effective, and addresses the concerns of various members within the utility
by providing the following advantages: a virtual wall to protect the individual
systems within each department from accidental or deliberate tampering by the end-user; allows data contributors to filter data and remove operational errors
before it can be viewed by a wider audience, thus minimizing false alarms and
maintaining departmental boundaries; and,
a single enterprise database eliminates concerns about proprietary software integration by converting all of the
data to a standard database protocol. Includes figures.