ECONOMIC ASPECTS OF COLD FOOD PASTEURIZATION

Proceedings of the 1999 Particle Accelerator Conference, New York, 1999
ECONOMIC ASPECTS OF COLD FOOD PASTEURIZATION
*
S. L. Bogart, Photon-Electric Pasteurization Corporation (US) , N. G. Tolstun, NIIEFA Delta (RF)
Abstract
X-ray machines come in “unit sizes” as they are de-
signed for a certain maximum power. Radioisotope ma-
The economics of cold food pasteurization are governed
chine power is a function of the inventory of radioactive
by a number of factors, including: the type of ionizing
material (e.g., Co-60) which can be adjusted in facility
radiation source (X-Ray, or Gamma), cost and power of
design to meet the specific throughput requirements of a
the source, pasteurization dose, location of the pasteuriza-
food manufacturing facility. Thus, to meet the needs of a
tion facility, facility capacity factor, and the annualized
specific facility, an X-ray machine may operate at less
costs. Using a costing procedure developed by Morrison
than its rated power output. This fact affects economic
[1], calculations for typical sources, locations, etc., have
comparisons unless care is taken to fit the current accel-
been updated using facility cost estimates prepared by a
erator technologies with the market needs.
major U.S. construction firm and scalings from the Morri-
son data.
Finally, irradiation plant scale and capacity factor affect
food irradiation economics. Morrison showed that econo-
The “owner’s cost” per pound of product is a function
mies of scale essentially vanish after a plant processes on
of the facility scale, showing an asymptote at ~ 100 kT
the order of 200 million pounds per year. We restrict our
(220 million pounds) per year of product. Likewise, the
analysis to plants for which economies of scale should not
owner’s cost significantly depends on the annualization
be in effect. We capture plant capacity factor with two
interest rate. A “stand-alone” location of the pasteuriza-
scenarios – 3 and 2 shift per day operation. The former
tion facility has an effect on the cost of the process due to
fully utilizes the capital equipment. The latter is better
the need to transport product from the meat plant to the
suited to most meat plants in the U.S.
facility, the labor for unloading and loading the product at
the facility, and the unshared G&A costs at the facility.
2 THE MODEL
This increases processing costs and “borrows” into the
shelf-life value of the product. An “integrated” location of
We use the costing formalism developed by Morrison in
the pasteurization facility (at the meat plant) minimizes
1989 and updated in the recent “Proposed Rules” [2] pub-
processing cost (inclusive of special labeling) and best fits
lished by the USDA for “Irradiation of Meat and Meat
the operating characteristics of a typical meat plant (2
Products.” For “sources,” we use X-rays and Co-60 with
shift, 5 day week).
cost estimates for the former from a survey in 1996 and
for the latter, from Morrison. For the “balance of plant”
1 INTRODUCTION
(including shielding, machinery, related expenses, escala-
tion, fees, etc.) we use a hybrid of costs – some scaled
Food irradiation has been thoroughly demonstrated to sig-
from Morrison and others from estimates of a US Engi-
nificantly reduce food contamination by pathological or-
neering & Construction firm. While the model is complex
ganisms by five orders of magnitude or more and extend
due to this “hybridization,” care has been taken to prop-
shelf-life. Irradiation may be performed by: electron-
erly allocate costs and escalation expenses.
beams, radioisotope gamma decay, and X-rays. The first
is the most efficient in terms of energetics but has serious
The source characteristics are presented in Table 1. We
limitations on product thickness, homogeneity (e.g., need
treat three sources – a low-cost/high-power accelerator, a
for boneless product), and packaging. The second is the
high-cost/low-power accelerator, and Cobalt-60. As a spe-
“standard” of the irradiation industry. The third has been
cial case, we also assess the effect of the high-cost accel-
made possible by recent advancements in electron-beam
erator operating at 7.5 MeV versus the current limit of 5.0
source power sizes, costs, and energies.
MeV. For Cobalt-60 irradiators, we “scale” the facility
with the Cobalt loading required to meet the throughput
We analyze the radioisotope and X-ray options as we
and assign a nominal price for the source material at
conclude that only these meet the current and future re-
$1.20/Ci.
quirements of the food manufacturing industry – pene-
trating, uniform, and cost-effective irradiation of food
The case parameters presented in Table 2 correspond to
products in final shipping boxes. Electron-beams are lim-
the two highest through-put Morrison cases (208 and 416
ited to product thickness of ~3 inches in comparison with
million pounds per year for poultry). For an “integrated”
tens of inches for X-rays and Gamma rays.
plant (irradiation performed at the meat manufacturing
facility), the lower of the two throughput cases (190
Tonne/shift) would meet the needs of ~ ½ of the ground
* - elbogart@aol.com
meat production facilities in the U.S. For the higher
0-7803-5573-3/99/$10.00@1999 IEEE.
603

---

Proceedings of the 1999 Particle Accelerator Conference, New York, 1999
throughput level (380 Tonne/shift), there is no known
facilities. The “stand-alone” plants have lower hourly
plant in the U.S. requiring this production level at this
throughput rates as they run three versus two shifts for the
time. Thus, such an integrated facility would be expected
same annual production rates. On a shift basis, they are
to irradiate other beef products as well. For poultry, the
better sized for a large fraction of U.S. meat manufactur-
low throughput rate for the integrated plant would meet
ers.
the needs of virtually all of the current U.S. manufacturing
Table 1 – Irradiation Source Parameters
Source Parameters
Low-Cost Accelerator High-Cost Accelerator
Radioisotope Co-60
Unit Power (kW)
500
200
As Req.
Unit Cost (1000 $ or $/Ci)
2000
4000
1.2
Unit Cost/kW of E-beam (1000$/kW)
4
20
NA
Unit Cost/kW of X-ray/Gamma-ray (1000$/kW)
60
240
80
Net Utilization Efficiency
0.4
0.4
0.25-0.4
Unit Cost/kW of X-ray: 7.5 MeV (1000$/kW)
155
Table 2 – Case Parameters
Sources
Balance of Plant
Notes
X-Ray
Based on Current A&E Costs
Based on Accelerator Unit Sizes
Based on Product Throughput
Radioisotope (Gamma)
Scaled from 1989 Morrison
Based on Product Throughput
Based on Product Throughput
Plant Type
Stand Alone
Low Rate
High Rate
225 to 450 Million Pounds per year
(Three 7 Hour. Shifts)
(Tonne/Shift)
(Tonne/Shift)
Adders to Irradiation
130
260
0.2 Cents each for transportation and labeling
Integrated
215 to 430 Million Pounds per year
(Two 8 Hour Shifts)
190
380
Adders to Irradiation
0.2 Cents for labeling
Major Cost Variables
Annualization
Interest Rate (%)
5 to 15
5 to 15
All Cases
Dose (kGy)
2 to 3
2 to 3
Range for Selected Cases – 2.5 Nominal
Cost of Cobalt ($/Ci)
1.2 to 1.5
1.2 to 1.5
Range for Selected Costs – 1.2 Low
As noted in Table 2, the “source” scaling is based on
Costs for transportation (stand-alone plant) and labeling
integer accelerator sizes and continuous Cobalt-60 re-
(both plants) were taken from the USDA promulgation of
quirements. This means that, for Cobalt-60, the radiative
the proposed meat irradiation rules – $0.002/lb each. We
power will exactly meet the selected processing rate re-
accept these estimates, but note that they will be product
quirements but, for the accelerators, the available radiative
and process-specific. We also note that no costs were pro-
power will always exceed the required power for the se-
vided for the development and production of packaging
lected processing rate – e.g., the capacity factor for the
materials that may be required for irradiation. We exam-
accelerator plant will be less than one for both integrated
ine the parametric variation of unit costs ($/lb) as a func-
and stand-alone plants.
tion of “interest rate” used to calculate the annualized
costs. Morrison assumed 5% which we felt was low in
Costs for the X-ray plant “balance of plant” were devel-
comparison with the food industries’ required rate of re-
oped from estimates prepared in 1998 by a U.S. architect
turn. An interest rate of 15%, the upper bound, still may
& construction firm and are linearly scaled according to
be too low. Finally, our model includes the capability to
plant hourly throughput. For the Cobalt-60 plant, we used
assess the effect of dose and Cobalt costs, etc., on overall
the 1988 Morrison estimates escalated to 1999 at 4 percent
costs. We do not parametrically assess these effects, but
per year and, again, linearly scaled according to plant
use them to examine special cases.
hourly throughput. For the plant processing rates analyzed,
these linear scalings are justified as there were no econo-
mies of scale (capital cost) above Morrison’s 208 million
3 RESULTS
pound/year throughput.
Table 3 presents the parametric results for both capital
investment and unit costs. For the former, it is evident that
604

---

Proceedings of the 1999 Particle Accelerator Conference, New York, 1999
the flat economies-of-scale prevail in the stand-alone plant
due to a better fit of integer accelerators with plant re-
and, to a lessor extent, in the integrated plant – the latter is
quirements at higher throughput.
Table 3 – Parametric Analysis Results
Plant TypeÍ
Stand Alone Plant (3 shift/day)
Integrated Plant (2 shift/day)
Low-cost
High-cost
Radioisotope
Low-cost
High-cost
Radioisotope
Accelerator
Accelerator
Cobalt-60
Accelerator
Accelerator
Cobalt-60
CostsÏ
130 T/Shift
190 T/Shift
Investment (Million $)
5.74
12.16
8.50
8.47
16.97
10.64
Total Cost (Cents/lb)
Total Cost (Cents/lb)
Annualization Rate – 5%
1.13
1.59
1.41
0.96
1.56
1.33
Annualization Rate – 10%
1.23
1.78
1.55
1.10
1.84
1.51
Annualization Rate – 15%
1.33
2.00
1.71
1.25
2.15
1.71
260 T/Shift
380 T/Shift
Investment (Million $)
11.48
24.31
17.00
14.67
29.33
21.27
Total Cost (Cents/lb)
Total Cost (Cents/lb)
Annualization Rate – 5%
1.03
1.48
1.30
0.81
1.33
1.26
Annualization Rate – 10%
1.12
1.67
1.44
0.93
1.57
1.44
Annualization Rate – 15%
1.22
1.89
1.60
1.06
1.84
1.64
In contrast with intuition, the stand-alone plant has
Last, a case was run to assess the effect of higher dose
poorer performance than the integrated plant. This is due
(3.5 kGy) for an interest rate of 10%. For the twelve cases
to the relatively low annualized irradiation costs in com-
at 10% in Table 3, the average increase in unit cost was
parison with the transportation cost “adder” for the stand-
less than 20% for the 40% increase in dose. This is due to
alone plant versus the lower annual costs for the integrated
the significance of the costs unrelated to the irradiation
plant from reduced manpower.
process – e.g., labeling and transportation.
The interest rate for annualizing costs is seen to have a
significant effect for the higher cost plants. However, the
4 CONCLUSIONS
overall unit costs for all cases are quite small and the ex-
Irradiation costs in general:
trema differ by only ~ 1.2 cent over the parametric range.
• Annualized irradiation unit costs are very low in
The low-cost, high-power accelerator offers the minimum
comparison with product production costs. However,
cost, followed by Cobalt-60, with the high-cost/low-power
the cost differences between technologies can be $1-2
accelerator being the highest.
million/year;
A calculation was made for the high-cost/low-power
• Packaging and transportation costs significantly add
accelerator at 7.5 MeV for product throughputs that re-
to irradiation costs. Other costs, such as oxygen con-
sulted in a nearly integer number of accelerators at an
trol, may increase unit costs;
interest rate of 10% (midrange) to determine if this higher
• Capital cost differences may be the major selection
energy operation had a significant effect on unit costs. For
discriminator because of low unit costs;
the stand-alone plant, the unit costs were 1.54 and 1.41
• Integrated irradiators have lower unit costs because of
cents/lb for processing rates of 104 and 208 Tonne/shift,
lower transportation and manpower costs.
respectively. For the integrated plant, the unit costs were
Irradiation costs for specific technologies:
1.38, 1.25, and 1.19 cents/lb for processing rates of 119,
238, and 476 Tonne/shift, respectively. This resulted in
• The least-cost technology is the low-cost accelerator,
costs that were competitive to Cobalt-60 for the stand-
followed by Cobalt-60 then the high-cost accelerator;
alone plant and less expensive than Cobalt-60 for the inte-
• X-ray machine selection should be made to meet the
grated plant (but still more costly than the low-cost/high-
processing requirements of meat production plants;
power accelerator).
• Irradiation environment optimization should be per-
formed for all irradiation technologies.
A special case was run to assess the effect of a higher
Net Utilization Efficiency (40%) for Cobalt-60, based on
reports that near 40% had been demonstrated in existing
5 REFERENCES
plants. At a 10% interest rate, this reduced the costs of

[1] Morrison, R.M. “An Economic Analysis of Electron Accelerators
Cobalt-60 irradiation to 1.32 and 1.21 cents/lb for the
and Cobalt-60 for Irradiating Food,” USDA, TBN 1762, 1989.
stand-alone plant and 1.17 and 1.10 cents/lb for the inte-

grated plant. These are significant improvements.
[2] Proposed Rules, USDA, Food Safety and Inspection Service, RIN
0583-AC50, “Irradiation of Meat and Meat Products.”
605

---