(190b) Connecting the Dots – Gaining Insight from Big Data in Hydrocarbons Manufacturing | AIChE

(190b) Connecting the Dots – Gaining Insight from Big Data in Hydrocarbons Manufacturing


Luijten, S., Dow Benelux B.V.
Li, L., The Dow Chemical Company
McAdon, M., The Dow Chemical Company
Hamilton, J., Dow Chemical Co
In hydrocarbons manufacturing large and complex data sets that elude traditional processing applications are rather common, and in an industry as established as olefins manufacturing the challenge to extract information from the data is not a recent one. Process and analytical data are collected continuously over years and stored on various data historians. Within the data, the operational complexities require intimate process knowledge to segregate planned start-ups and shut-downs, unexpected process events, trials and normal operation periods. In addition, there is significant inter-dependency of the various process variables. However, through better computing power and the availability of advanced modeling and visualization platforms, it is possible to extract new insights about plant performance, multivariate relationships, conduct more reliable forecasting and enable more robust real-time monitoring.

Dow’s Hydrocarbons business operates 14 plants with a couple of hundred furnaces. Outgrowing the practice of the past of evaluating single furnace data, opportunities from global performance assessment range from more robust furnace coil life time prediction and technology comparisons to the identification and prioritization of reliability issues and operational improvements. A system that collects data from different storage locations and data bases, and automatically performs time-alignment and data pre-processing was developed. Subsequently, visualization in a Tableau dashboard to allow the identification of dependencies and connections. It enabled understanding of KPI variations between furnace technologies as well as differences between different units. In addition, the data can be used to build streamlined models of the cracking and coking behavior for furnaces accounting for different operating conditions and technologies. Within Dow, this has allowed to improve the performance of ethylene furnaces in several aspects.



This paper has an Extended Abstract file available; you must purchase the conference proceedings to access it.


Do you already own this?



AIChE Pro Members $150.00
AIChE Graduate Student Members Free
AIChE Undergraduate Student Members Free
AIChE Explorer Members $225.00
Non-Members $225.00