Systems Informatics and Analysis - Engineering and Science of Biomass Feedstock Production and Provision - page 211

management and operational decisions. The model has been extensively applied to
study various scenarios [ 1 , 17 , 88 , 89 ]. The BioFeed model has been developed such
that users can select the equipment a priori and then optimize only their manage-
ment decisions. This allows the model to simulate very specifi c cases and thus
extends the scope of the applications. One of the important features of BioFeed is
the consideration of different farm sizes based on actual farm size distribution in the
Illinois, USA. This is quite important as Shastri et al. [ 1 ] showed that farm size
signifi cantly impacts the on-farm production cost.
Recently, Lin et al. [ 90 ] have developed a new optimization model, named
BioScope (Biomass Supply Chain Optimization). This model proposes that inter-
mediate centralized storage and preprocessing centers (CSPs) are essential to
improve the supply effi ciency of biomass feedstock, and optimizing their location is
critical. The model uses an MILP approach to optimize the location and size of
these CSPs as well as the biorefi neries. The model uses GIS-based information to
determine the potential biomass supply at county level and also employs GIS-based
transportation data to calculate road transportation distances. An important feature
of the model is that it considers the biomass supply and demand constraints over a
number of years and, therefore, provides the optimal strategy to develop the bio-
mass feedstock sector over a long time horizon (15 years). The model has been
successfully integrated with BioFeed, which provides the detailed farm-level pro-
duction cost estimates that are used by BioScope to perform simulations. BioTrAnS
(Biomass Transportation Analysis System) is another optimization model that is
currently under development by the same group, which optimizes the short-term
(hourly to daily) transportation and logistical decisions. The current focus is on
optimizing the dispatch timings of each truck for picking up biomass from farms or
storage facilities and delivering it to the destination in order to minimize the idling
time in queues for loading and unloading. It takes output from BioFeed, a strategic
level model, and further optimizes the short-term logistics decisions.
Leboreiro and Hilaly [ 91 ] developed a model to study the collection, storage, and
transportation of biomass and used it to optimize the biorefi nery capacity. For two
different scenarios with corn stover as feedstock, the optimal biorefi nery capacities
were 3,450 and 4,550 Mg d −1 and the optimal ethanol production costs were $0.45 and
$0.47 l −1 . Zhu et al. [ 92 ] have developed an MILP model that optimizes the strategic
decisions such as the locations of the biorefi nery and warehouses and tactical deci-
sions such as the transportation schedules. It covers the operations of harvesting, stor-
age, transportation, and biofuel production. The model uses monthly time steps for
decision making and 1 year as the simulation horizon. Zhu and Yao [ 93 ] modifi ed the
model to consider supply of multiple feedstocks and showed that the total profi t
increased by almost 50 % by using three different feedstocks instead of one. Sultana
and Kumar [ 94 ] also optimized the transport of a mix of biomass feedstocks and
determined that 30 % agricultural residue as bales and 70 % forest biomass as chips
led to minimum transportation cost for a biorefi nery of capacity 5,000 Mg d −1 . Other
optimization-based studies include Zuo et al. [ 95 ], Mapemba et al. [ 96 ], An et al. [ 97 ],
and Kim et al. [ 98 , 99 ]. Results for some of these studies are reported in Table 8.2 .