Customer Login >
Data Reconciliation
Robust and Accurate Data Reconciliation








dotted rule
What differentiates Inlibra’s data reconciliation algorithm
from other applications?

Other data reconciliation applications

logo bug  Operate on simple process models with limited flexibility for constructing
complex models.

logo bug  Wrongly incorporate values associated with gross errors, missing flows and invalid models into their reconciled results.

logo bug  Require manual re-execution after gross errors and modeling errors.

logo bug  Limit balancing work to either mass or volume but not both in a given run.

logo bug  Distribute losses and gains to meters throughout the model regardless of the size of the discrepancy.

dotted rule
Inlibra solves these problems differently. Most importantly, it automatically identifies and excludes gross errors from reconciled results. This can reduce the required number of iterations to as low as one.

How Inlibra operates:

logo bug  Validates model network problems before detecting any missing movements and /or faulty measurements.

logo bug  Detects, identifies and removes missing movements before detecting faulty measurements.

logo bug  Detects, identifies and removes all faulty measurements before the data reconciliation process.

logo bug  Runs data reconciliation while examining for gross errors to produce the best optimal reconciled values.

logo bug  Performs data reconciliation with mass and volume balances simultaneously.

logo bug  Permits configuration of “sanity check” thresholds to prevent distribution of large losses and gains.

dotted rule
This unique data reconciliation with mass and/or volume balance gives you the powerful flexibility of configuring process models according to the customers’ specific needs. For example, process unit areas are often required to reconcile with both mass and volume simultaneously, while tank farm areas are usually reconciled with only volume due to the inaccuracy of the density lab data measured in the tanks. Inlibra allows you to select balance types (mass and/or volume balance) for each process and tank.

Inlibra performs data reconciliation with the sparse matrix method. It takes only seconds to solve thousands of flows. It permits the user to configure thresholds for discrepancies at critical points in the model. Inlibra allows the user to set the minimum constraint for the loss so that it can be distributed into the measurement streams if the calculated loss is within the acceptable range. If the calculated loss exceeds the minimum constraint, it indicates the user must recheck the gathered input data.

Inlibra’s advanced mathematical approach is more than just an intellectual exercise. Inlibra provides users with many practical tools for solving everyday problems. Take, for example, the issue that often arises with utility units, where there’s a set of unmeasured steam flows whose overall flow can be inferred. The problem is that the flow for each of the individual unmeasured streams cannot be inferred. Inlibra, unlike competitor products, permits users to configure a ratio for each unmeasured flow that governs how much of the inferred total gets assigned to each. Practical features such as this have been incorporated into Inlibra for more than a decade, based on feedback from users.

dotted rule

Products >
Data Reconciliation
   Introduction >
   Robust and Accurate DR >
   Graphical User Interface >
Composition Tracking
   Introduction >
   Relationship of CT to DR >
   Technology Description >
Smart Objects
Stockpile Tracker
   Introduction >
   The Problem >
   What ST Provides >
   What Your Plant Needs >
   GUI Features >
   Goals of ST Design >
   Modeling Method >
   Security >

rule190Downloads (pdf)
Product Brochures
Data Reconciliation >
Composition Tracking >
Stockpile Tracker >
White Paper
What is Data Reconciliation? >
Case Studies >



















Back to Top >