Abstract: The invention relates to a process for producing nitrobenzene by an adiabatic nitration of benzene (2) and nitric acid (1) in the presence of sulfuric acid (7), the process comprising: (a) feeding benzene (2), nitric acid (1) and sulfuric acid (7) into a reaction zone (R); (b) producing a reaction mixture (4) comprising water, nitrobenzene and sulfuric acid by reaction of the benzene and the nitric acid in the presence of the sulfuric acid in the reaction zone (R); and (c) separating the reaction mixture into an aqueous phase (6) comprising water and sulfuric acid and an organic phase (5) comprising crude nitrobenzene by phase separation; wherein the concentration of titanium in the reaction mixture (4) is kept below 170 mg/kg.

Abstract: In order to predict the future evolution of a health-state of an equipment and/or a processing unit of a chemical production plant, e.g., a steam cracker, a computer-implemented method is provided, which builds a data-driven model for the future key performance indicator based on the key performance indicator of today, the processing condition of today, and the processing condition over a prediction horizon.

Abstract: The invention relates to a process for producing aniline by catalytic hydrogenation of nitrobenzene comprising: (a) reacting hydrogen and nitrobenzene in the presence of a catalyst in a hydrogenation reactor, forming a crude reaction product; (b) working up the crude reaction product to obtain pure aniline, wherein the composition of the crude reaction product is analysed continuously or at regular time intervals before starting working-up in (b), for determining the proportion of by-products, wherein, in the event of an increase in the amount of by-products, measures are taken to prevent nitrobenzene from breaking through in the hydrogenation reactor.

Abstract: In order to predict the future evolution of a health-state of an equipment and/or a processing unit of a chemical production plant, e.g., a steam cracker, a computer-implemented method is provided, which builds a data-driven model for the future key performance indicator based on the key performance indicator of today, the processing condition of today, and the processing condition over a prediction horizon.