“Honeywell’s decades of product development and industry expertise have helped us devise an exceptionallyeasy to operate and maintain solution that prepares us for the future in the most complex task of refineryplanning. As a result, not only is our new system running smoothly and more efficiently, we have also greatlyimproved internal work practices while also putting ourselves in a much more flexible position for adapting and reacting to the coming changes and mandates in our highly competitive market.”
Dr. Abbas Azizi, Linear Programming Specialist, Engen Petroleum Ltd.
Engen sought an innovative, yet simple-to-operate system for upgrading its overall refinery capability to improve production schedules and prepare for production of new products such as biofuels.
Among the benefits now possible from the use of Honeywell’s integrated system are:
Engen Petroleum Ltd. has a proud history that has held strong through wars, apartheid and recessions. Today, Engen is recognized not just as an industry leader, but for its stellar record in business, ethics, justice, corporate citizenship and equality.
A subsidiary of Malaysia's Petroliam Nasional Bhd (Petronas), Engen boasts the largest market share in South Africa’s petroleum industry and has interests in more than 20 African countries.. Engen owns and operates its own refinery, which produces more than 125,000 barrels of oil per day, employing about 3,000 permanent staffers.
Engen Petroleum’s interests include refining crude oil, marketing and retailing primary refined petroleum products and providing consumer convenience services through its retail network. The company owns and manages a complex crude oil refinery and a state-of-the-art lubricants blending plant at Durban.
The Durban refinery produces both lead replacement and unleaded gasoline as well as lubricants and chemicals,. With the increasing demand for unleaded gasoline, Engen needed to find a solution to optimize the volume usage of the high-octane components, which would facilitate a refinery-wide move toward a lead-free gasoline market. With this in mind, Engen sought input for the design of a new system. The company’s vast experiences in the refinery business convinced executives that an entirely new approach was needed to design and implement this system, so the result could be run easily by refinery planners and schedulers. This approach necessitated the maximum involvement of internal resources and staff expertise with the minimum reliance on the external resources.
As Engen’s relationship with Honeywell dates back several decades, it was logical for the company to call upon the industry experts to provide their view on constructing, implementing and operating such an all-encompassing, state-of-the-art solution. To their mutual benefit, Engen has vast, documented experience with its previous planning and scheduling system, so there was much historical data to call upon when formulating its replacement. This information would come in handy to tackle the high level of complexity involved in building the proper system to optimize the gasoline and bunker fuel oil blending operations.
Stellar data integration as well as system performance and functionality were some of the key issues that had to be tackled with system design. Engen also demanded a system that would be simple and efficient to operate so refinery schedulers would not require additional resources in order to maintain daily routines.
“Any change in fuel specification affects the entire supply chain,” said Dr. Abbas Azizi, Linear Programming Specialist, Engen Petroleum Ltd. “This process requires a complex mathematical model to take into consideration a number of factors, including production by site, raw materials, product components and cost/pricing; product manufacture, blending and specification.”
As Engen saw the system, the task at hand required meeting a list of key functional objectives as well as a number of crucial systems design requirements while adhering to clearly defined implementation guidelines. The final goal was to transition from having an acceptable plan to an optimal – or close to optimal – plan.
Some of the objectives of this project included:
To ensure that an optimized plan is converted to an optimized schedule, Engen utilized an integrated planning and scheduling system, where the refinery-wide linear programming (LP) is linked to a multi-period product blending and scheduling LP system. Engen Petroleum is known internationally as one of the pioneers in the use of Linear Programming (LP) techniques for plotting and scheduling all planning aspects within its refinery planning. To support this imperative function; Engen traditionally has relied upon Honeywell’s Refinery & Petrochemical Modeling System (RPMS) to integrate with the responsibilities of refinery planning. Recently, as the move toward cleaner fuels gains continued momentum in Africa, the system is also used to study the feasibility and economics of various lead phase-ou strategies.
In use at Engen since 1989, RPMS is linked to Honeywell’s ASSAY system for gathering and compiling crude assay data. What Engen had desired for a long time was a specific tool to support the optimization of its product blending and scheduling functions.
The company’s blending operations involve the pooling of various gasoline components into tanks and the isolation of other components into multiple tanks dedicated to key components. These components then need to be blended under online blend control to dedicated product tanks.
The new system supports this capability, where the supply planning model provides input to the monthly refinery operating planning process. Generating the monthly Refinery Operating Plan (ROP) is one of the major applications of the new Honeywell system. The system is used to satisfy a multi-period (three-month roll-over) refinery production plan. The results of the ROP provide input to the multi-period blending and scheduling process. Then the final application electronically passes the blending orders to the online blend control system.
Monthly planning involves the usual supply chain activities, including crude evaluation and selection, process optimization, product manufacture for a segmented market as well as product inventory and distribution. To meet the diverse needs of a broad set of business line functions, the integrated system essentially works as a decision support tool connecting the various activities derived from these “information silos” and passes on this data for use in monthly planning exercises.
Data from the various sources are routed to the refinery-wide LP. The data from the refinery optimized plan then is routed to blend the scheduling LP in the plant operations department as well as the advanced control systems. The run-down tank volume and qualities are regularly updated via the refinery-wide database using Honeywell’s Uniformance PHD historian. Honeywell’s RPMS LP uses Xpress from Dash Associates for linear and mixed integer programming. The same solver is used in BLEND scheduling for LP as well as MIP related blending cases..
“As we look to the future, one requirement we are planning for is the need to move continually toward cleaner burning fuels,” says Dr. Azizi. “South Africa is no exception. Our first emphasis will be to reduce the sulfur in both our gasoline and diesel products. The second emphasis will be to understand the impact of increasing the production of unleaded gasoline. Honeywell’s RPMS will play an integral role in making this process a success.”
Investigating both these issues creates a challenging role for the newly introduced integrated planning and scheduling system. The goal for Engen is to maximize gasoline production while meeting the lower sulfur and the higher octane demand.. Considering that the octane requirements in South Africa are higher than in many other parts of the world, the move toward more unleaded gasoline with the lower sulfur and the higher octane creates a considerable octane imbalance.
Due to the number of factors involved with the new platform, the addition of various new units will be undertaken in a series of modeled-in phases. Although this approach may be timeconsuming, it will allow Engen to gain a better understanding of the underlying mechanism driving the complex mathematical model.
Concludes Dr. Azizi, “As the production of new-generation products, such as biofuels, gain impetus at Engen, new mathematical models, and in particular the gains in linear programming techniques gained from our relationship with Honeywell, are proving to be powerful decision science system for studying the impact of complex problems such as bio fuel production.”