Öner Hortaçsu

General continuous time models for production planning and scheduling of batch processing plants: mixed integer linear program formulations and computational issues

Abstract In this work, a continuous time model for optimal planning and scheduling of the production in batch processing plants is developed. The considered plants are general in the sense the products can run through different processing stages and follow different manufacturing routes. Processing stages are represented as operations. They can be viewed as modules of operators accomplishing same tasks and sharing possibly different operational characteristics. The resulting continuous time mixed integer nonlinear program (MINLP) is capable to handle complicating situations such as batch splitting, resource allocation and equipment maintenance. By using known linearization techniques, the MINLP is reformulated as a mixed integer linear program. It is further refined by using a modified version of the reformulation linearization technique and some other equivalent reformulations. The model is also implemented on a real life case: paint production. The computation of optimal production plan and schedule takes only a few minutes for this real case.

Scheduling of batch processes with operational uncertainties

Abstract In this work, a mathematical programming model for optimal scheduling of the operations of a batch processing chemical plant is developed. The model is capable to handle all possible deterministic variations in the set-up and operation times of batch operations, and the model is sufficiently general to include the uncertainties introduced by the probabilistic behavior of set-up and operation times of batches when such variations can be mathematically defined. It is shown that the probabilistic model can be reduced to a Mixed Integer Non-Linear Program (MINLP), once the probability distribution functions used to model the random variations are defined. The resulting MINLP is then linearized and solved for small examples using CPLEX 3.0 in order to obtain a schedule which maximizes the net expected operational profit.

Multi-component models for seed and essential oil extractions

Abstract Various types of models have been presented for the SFE of seed and essential oils in the literature. Most of these models consider the natural matrix as a porous sphere and the extractable material as a single chemical specie. In this work, the shrinking-core model is modified for application to a multi-component system in order to simulate SFE of seed oil components. The porous sphere model with the desorption–dissolution–diffusion mechanism to express the extraction is also extended for application to a multi-component system to simulate the SFE of essential oils. The results of these more rigorous analyses suggest possibilities for selective extraction of the mixture constituents, if some processing parameters could be better controlled. Selective extraction would eliminate the need for additional separation operations, which would reduce the cost of production of the desired target component(s).

Semi-batch deterpenation of origanum oil by dense carbon dioxide

Abstract Semi-batch deterpenation of origanum oil (Origanum Munituflorum) by dense (sub/supercritical) CO2 was studied in an unpacked column with two temperature zones which were applied to columns upper and lower sections. Experimental results at combinations of 70 and 85 bar pressure and 38 and 55°C temperature levels were reported. The results were discussed in terms of 14 major components, and the monoterpene (MT) and non-monoterpene (NMT) fractions of the oil. The oil was obtained from the top of the column at four different deterpenation time-cuts (integral samples of 0–15, 15–25, 25–35, and 35–45 min time intervals). In addition, the effects of the amount of oil feed charged to the extraction vessel, the solvent-to-oil ratio, and the presence of glass beads (0.5 cm) in the column as packing material were demonstrated. The best separation between the MTs and NMTs was achieved at 70 bar with 38°C lower-section and 55°C upper-section temperatures. At these conditions, the MTs are preferentially separated (with relative-distribution-ratio values greater than 2.0) from NMTs (with relative-distribution-ratio values less than 0.5), almost independent of the deterpenation time (flat maximum occurs in the 15–25 min deterpenation time-cut). It is concluded that any occurrence of single-phase state in the column deteriorates the effectiveness of the separation. Lower densities (higher temperatures) at the upper section of the column enhance the separation by causing internal reflux. Statistical analyses of the data and a response-surface analysis based on empirical models are presented. The suggested processing scheme that employs selective CO2 extraction and rectification has the potential to be used in essential-oil refining technology.

Equilibrium distributions of key components of spearmint oil in sub/supercritical carbon dioxide

Effects of temperature (at 35, 45 or 55°C) and pressure (10–110 atm) on the relative distribution coefficients of the twelve key components of spearmint oil (essential oil ofMentha cardiaca; Scotch spearmint) at equilibrium in dense CO2 were investigated under conditions ranging from subcritical to supercritical regions. Effects of vapor pressure, molecular weight and polarity of the key components on their equilibrium distributions in sub/supercritical CO2 are discussed. At 35°C, all key components of spearmint oil are equally soluble in dense CO2 within the 12–102 atm pressure region. At 45 and 55°C, the key components are equally soluble for pressures greater than about 60 atm. However, around either 45°C/27 atm or 55°C/35 atm conditions, the relative distribution coefficients of all monoterpene hydrocarbons and of isomenthone (an oxygenated monoterpene) exhibit maxima, which are due to significantly higher vapor pressures of these components and significantly lower solvating power of the dense-gas solvent at these particular temperatures and pressures. Vapor-pressure effects, coupled with the decrease in solvating power, dominate the effects of polarity and molecular mass of the key components. Deterpenation of spearmint oil with dense CO2 is possible around either 45°C/27 atm or 55°C/35 atm, where the monoterpene hydrocarbons tend to concentrate in the CO2-rich phase.

Semi-batch packed-column deterpenation of origanum oil by dense carbon dioxide

Abstract The study aims at the separation of the undesired, light monoterpenes (MTs) of origanum oil from more valuable and heavier non-monoterpenes (NMTs). The scope of this work is the analysis of fundamental aspects of semi-batch, packed-column deterpenation process of origanum oil by dense CO2 at 70/85 bar pressure and 38/55°C temperature levels. Effects of operating pressure, temperature zones, presence of packing, and entrainer (ethanol) use are investigated. GC/MS analyses of 14 major components constituting the bulk (86.4%) of the oil samples obtained from the top of the packed column in four different deterpenation time-cuts show that the resulting deterpenated product is rich in MTs (48–89% MTs, 52–11% NMTs). The oil remaining in the extraction vessel is rich in high-molecular-weight NMTs (15–24% MTs, 85–76% NMTs). Thus, the quality of origanum oil is improved as compared to the oil charged (36.44% MTs, 63.56% NMTs). Separation deteriorates with increasing deterpenation time. Enhancement of separation with packing is observed at the first time-cut. No significant entrainer effect (1.25, 2.5, and 5% ethanol in the charge) is noticed. Statistical analyses of the data and a response-surface analysis based on empirical models are presented. Suggested processing scheme that employs selective CO2 extraction and rectification has the potential for use in essential-oil refining technology.

Scheduling of batch processes : An industrial application in paint industry

Abstract In this work, the mathematical programming model for optimal scheduling of the operations of a batch processing chemical plant, capable to handle all possible deterministic and many stochastic variations in the set-up and operation times of the batch operations developed previously (Orcun et al. , 1996, 1995, and 1994) is applied to operations in a plant of a leading Turkish paint manufacturer. The plant focused on in this work is a multi-product, batch processing plant having a wide variety of products, competing for various process equipment on a production site. The plant under consideration operates with respect to an “order based” production policy. Thus, this application can be classified as a short term scheduling of a real case multi-product batch processing plant. The MILP (Mixed Integer Linear Programming) formulation generated as a result of the general modeling framework developed, is solved using the CPLEX 3.0 to obtain schedules maximizing the net operational plant profit under varying market conditions. Moreover, the analysis of the schedules obtained from the solution of the problems formulated yielded some information on the performance, utilization and organization of the production site.

Optimal planning and scheduling of batch plants with operational uncertainties: An industrial application to Baker's yeast production

Abstract In this work, the mixed integer linear programming (MILP) model developed in Orcun et. al 1996 for optimal planning and scheduling of batch process plants under uncertain operating conditions is further improved to deal also with discrete probability functions. Furthermore, the logic behind integrating the processing uncertainties within the MILP model is implemented on the variations in the production volumes that can be faced in some batch processes such as Bakers yeast production. The modified model is tested on Bakers yeast production plant data to illustrate the effect of uncertainties on the production planning and scheduling. The results show that the plant production will be improved by 20% when the optimal production planning and scheduling is utilized by fine tuning the degree of risk the management can resist. An example on how a process design engineer may utilize such an MILP model for optimal planning and scheduling of batch process plant and identify plant problems, such as the bottleneck operations, is also included. A simulation type analysis on how to improve the processing site, i.e. the effect of introducing an extra operator to the bottleneck operation, is also demonstrated in this work using the available plant data.

Double-bond depletion of soybean oil triglycerides with KMnO4/H2O in dense carbon dioxide

Soybean oil triglycerides (SOT) can be used for the synthesis of rigid polymers. This research investigates the potential of using dense (sub/supercritical) CO2 in the reaction medium for the addition of functional groups to SOT. As an alternative and novel method, the reaction of SOT with KMnO4 in the presence of water and dense CO2 is presented. Dense CO2 is utilized to bring the soybean oil and aqueous KMnO4 solution into contact. Experiments are done at 10,25, 34.5, 50 °C and 2.5, 5, 7, 11, 16 MPa. Effects of temperature, pressure, NaHCO3 addition, and KMnO4 amount on the conversion (depletion by bond opening) of soybean-triglyceride double bonds (STDB) are investigated. The highest STDB conversions, about 40%, are obtained at the near-critical conditions of CO2. The addition of NaHCO3 enhances the conversion; one mole of NaHCO3 per mole of KMnO4 gives the highest benefit. Increasing KMnO4 up to 10% increases the conversion of STDB.

Baker’s Yeast Plant Scheduling for Wastewater Equalization

A Baker’s yeast plant was modeled using data obtained from the normal operation of the plant via Monte Carlo analysis. The model which was first validated with respect to the normal plant operations was then used to predict wastewater discharges from two scheduling scenarios. The results showed that the plant wastewater discharge may be equalized if a 25% reduction may be considered to be economically acceptable since wastewater equalization will not necessitate additional facilities for wastewater equalization external to the existing. However, if only one fermentation vat may be added to the process then wastewater equalization would be realized with 24% increase in the capacity. The additional cost of the new vat is considered to be of set by the increased plant capacity.