Optimal asset management algorithm within 20 – 30 years period is proposed for enterprises that
have large tangible and intangible assets’ fleet including buildings, constructions, facilities, land
parcels, constructional projects, equipment units, software systems, data bases, etc. As an optimization
criterion a maximum total discounted profit for the period is selected. The development strategy
which will be selected must satisfy various constraints, reflecting the requirements of customers,
the environment, the regulators, the company’s owners and management. These constraints may
be financial, technical, market, environmental and other. The entire set of possible projects for the
acquisition, construction and commissioning of new assets, renovation and decommissioning of the
existing assets, as well as projects for the repair of assets and compliance with restrictions, actions
and maintenance works is divided into a collection of specially formed portfolios. Inside, all the elements
of the portfolios are ranked. Budgets of formed portfolios serve as control variables. The algorithm
involves the generation of variants of long-term development of the enterprise and iterate
over them in order to find a maximum of the total discounted profit for the whole period. Variants
determine terms of projects’ completion for the acquisition and construction of new assets and reconstruction
of existing ones. Then they are supplemented by the information about projects and
activities of maintenance and repair as well as constraint satisfaction. Each variant is divided into
yearly intervals, for which optimal internal supply chain plans are calculated. Feasibility assessment
of calculated plans is produced using simulation models, special models, models and methods
of predictive analytic. The architecture of the software system that implements the developed
algorithm is proposed.
Key words
optimization criterion, constraints, project and activity portfolios, internal supply chain planning, models of plans’ feasibility assessment, asset master data.