Decision Support System of Warehouse Allocation using Analytical

Hierarchy Process Method

Dina Fitria Murad

, Meta Amalya Dewi

, Yuvita Dyah Ayu Novitasari

, Aditya Soleh

and Hudaiﬁ

Information Systems Department, BINUS Online Learning, Bina Nusantara University, Jakarta, Indonesia

Information System Department, Tanri Abeng University, Jakarta, Indonesia

Keywords:

Decision Support System (DSS), Analytical Hierarchy Process (AHP), Warehouse Allocation.

Abstract:

This study aims to improve business processes by developing an innovation that can support the determination

of warehouse allocation decision making. The delay in determining the decision to allocate the warehouse is

the main reason for this research. The analysis process until the distribution of information is not systematic

so that it slows the distribution process. This research develops an information system that can automate the

determination of warehouse allocation decisions using Analytical Hierarchy Process (AHP) data processing

methods that can be sorted by the ranking of the proposed alternatives. The results showed that the solution

could accelerate the process of warehouse allocation so that the distribution process runs more optimally.

1 INTRODUCTION

The rapid development of today’s technology invites

the attention of various industrial sectors to utilize in-

formation system technology. Information systems

have become one of the basic needs that support al-

most all company activities. No exception in trad-

ing companies, information systems will signiﬁcantly

assist them in the process of distributing goods to

consumers(Dennis et al., 2012). A critical process

in a company is the decisionmaking process. Mak-

ing decisions is part of the process of considering,

understanding, remembering and reasoning about ev-

erything Decisions are taken by knowing and formu-

lating the problem clearly, then solving the problem

must be based on the selection of the best alternative

decisions. Because of the importance of this deci-

sionmaking process, which decisions must be taken

quickly and accurately, the Decision Support Sys-

tem (DSS) information system appears. This decision

support system is an information system which helps

company management to be able to make decisions

more quickly and accurately. This system combines

various data in the company, calculates and processes

it soon until ﬁnally, it produces information that will

help in the decision-making process. This company

has 3 leading warehouses spread across the Jakarta-

Tangerang area, namely in Cimanggis, Jatake (ﬁg.1),

Jatiasih (ﬁg.2)

space

Figure 1: Distribution to Jatake

Figure 2: Distribution to Jatiasih

The allocation process is still running using What-

sApp (ﬁg.3) assistance and occurs outside the analyst

distribution work hours, which is at 6 pm. So, it can

be said that the running process can take up time be-

cause the distribution of analysts takes approximately

7-10 minutes to analyse the data. Besides that, the use

of Microsoft Excel in analysing data manually allows

for human error. On the other hand, the destination

warehouse decision information must be fast and ac-

curate so that PIC Shipping is immediately followed

up and distributed to drivers for delivery. The slower

the ﬂow of information will undoubtedly affect the

294

Murad, D., Dewi, M., Novitasari, Y., Soleh, A. and Hudaiﬁ, .

Decision Support System of Warehouse Allocation using Analytical Hierarchy Process Method.

DOI: 10.5220/0009909302940298

In Proceedings of the International Conferences on Information System and Technology (CONRIST 2019), pages 294-298

ISBN: 978-989-758-453-4

speed of the process of distributing goods. This will

ultimately have an impact on the company’s business

processes, so companies need a system that can sup-

port the process of analysing the allocation of ware-

house destinations quickly and precisely.

Figure 3: Allocation process using WhatsApp

This is where the role of decision support sys-

tems is expected to help the organization. With the

adoption of a decision support system, the data analy-

sis process to determine the destination warehouse is

scheduled to run more quickly and systematically and

can reduce the burden of analysis on analyst distribu-

tion. With this author raised the idea to Analysis and

Develop Information Systems Supporting Warehouse

Allocation Decisions using AHP (Mu and Pereyra-

Rojas, 2016)(Saaty, 2008). AHP methods related to

DSS have been widely used and proven to provide

beneﬁts for companies (Susanto et al., 2017).

2 LITERATURE REVIEW

2.1 Decision Support System

Decision Support System (DSS) is a computer-based

decision support tool to assist decision-makers by pre-

senting information and interpretations for various al-

ternative decisions (Pal and Palmer, 2000). The aspect

of the scope system in DSS is automation which helps

decision-makers with different levels of intelligence.

From a theoretical point of view, decisions relate to

cognitive concepts, especially those related to ideas

that support humans in decision making. SPK can

help decision-makers in building strategic decisions.

The use of DSS has shown satisfactory results, by

minimising costs, accelerating the decision-making

process, and signiﬁcant achievements in competitive

advantage.

2.2 Analytic Hierarchy Process

The Analytic Hierarchy Process (AHP) developed by

Professor Thomas Saaty in 1980 made it possible to

arrange decisions hierarchically (to reduce their com-

plexity) and show the relationship between goals (or

criteria) and potential alternatives (Mu and Pereyra-

Rojas, 2016), the stages carried out refer to the ana-

lytic hierarchy process:

1. Develop a model for decisions: Break the decision

into a hierarchy of goals, criteria, and alternatives

2. Lower the priority (weight) for the criteria

3. Consistency

4. Lower local priorities (preferences) for alterna-

tives

5. Decrease Overall Priority (Synthesis Model)

6. Do a sensitivity analysis

7. Make Final Decisions.

2.3 Previous Research

The Yager Fuzzy MADM model chosen as the Yager

analysis model to determine the most prospective cus-

tomers with better feasibility through the highest giv-

ing(Susanto et al., 2017). By using the vector D value

that has been calculated from the sum of each cri-

terion value {0.904; 0.794; 0.914; 0.794}, the most

prospective customers are customers C (D3) with a

value of 0.914.

The customer is chosen as the most prospective

customer to be given Murabahah ﬁnancing. Each cri-

terion is given a weight value according to the desired

priority. The system that was built streamlined the

decision-making process time to be more efﬁcient.

DSS generates valuation calculations based on sub-

jective and objective assessment criteria for using 5C:

Character, Capacity, Capital, Guarantee and Condi-

tions. The system can present an assessment based on

the calculation of the value of each criterion such as

very good with a value of 1, 0.8, 0.6, 0.4 and 0.2 to

improve the weight vector results obtained from the

input value of the comparison criteria to produce the

most recommended customer data sequence. The suc-

cess of AHP on DSS was also carried out by other

researchers with different achievements according to

their individual needs (Dweiri et al., 2016) (Sari et al.,

2017)(Narabin and Boonjing, 2016).

Another researcher is by (Sinesi et al., 2017) gen-

erating DSS with A multivariate logic to monitor in

real-time for one week using GPS data from Marine-

Trafﬁcwebsite and validating the proposed model

by comparing the results of the model with real

data which counts four performance indicators (ME,

MAD, MAPE, MSE), the accuracy results are better

than the proposed model in evaluating the probability

of a maritime company choice.

Decision Support System of Warehouse Allocation using Analytical Hierarchy Process Method

295

From the literature that the author studied before,

there are some similarities and also differences. The

equation is that the above writing together aims to cre-

ate solutions in determining decisions of the issues

raised, whereas the difference is in terms of method-

ology and technique.

From the literature above, it is explained some use

the Fuzzy, Fuzzy MADM and AHP methods. From

the differences and similarities above, the author can

conclude that the research the writer raised is sim-

ilar to the 5th study(Narabin and Boonjing, 2016),

which together use the AHP methodology for deci-

sion making. This is because decision making has

multi-criteria decision making, so ranking techniques

are needed to determine the best. From our observa-

tion, the AHP methodology is judged to be following

the present case.

The AHP methodology itself was developed by

Thomas Saaty, in which decision making is calcu-

lated by placing priority scores in the form of a criteria

matrix, then calculating choices, and ﬁnally getting a

percentage of each option. The best choice is the pri-

ority score with the highest percentage. The authors

chose the AHP method because it has a clear hierar-

chical structure so that it will be able to reduce com-

plexity and be able to show a clear relationship be-

tween the criteria and alternative solutions proposed.

3 RESULT AND DISCUSSION

Of the 11 stages of the business process running, the

authors ﬁnd 8 steps that allow the process to be max-

imized. And, we found 6 stages of the process that

runs the system running namely:

1. In the PIC Shipping WMS interface, shipping data

is mixed between those that have not been allo-

cated, those that have been awarded, or those that

have been veriﬁed by the warehouse. And not

grouped by expedition ﬂeet police and plant num-

ber.

2. The waiting time for submitting summary ship-

ping list information from PIC Shipping to analyst

distribution is still quite long, which is around 2-3

minutes. This happens because the delivery of in-

formation is still going on using the help of What-

sApp

3. The analysis process always takes about 7-10

minutes because it is done manually by the dis-

tribution analyst utilising the support of a simple

application, thus allowing input errors or analysis

errors.

4. There is still a dependence on the presence of An-

alysts Distribution. So if the distribution analyst is

not carrying a smartphone (to check WhatsApp),

it will hamper the distribution process.

5. There are 2 shipping shifts from AHM, while

there are only 1 analyst distribution staff, so the

workload of 1 distribution analyst can reach 15

hours per day, in 5 working days.

6. There are times when there is still a suboptimal

balance of stock in each warehouse, due to uneven

allocation.

The stages of using AHP techniques in this study

are as follows:

1. Modelling

Here the criteria are determined which are used as

a reference for problem analysis. Measures that

have been obtained from the results of interviews

with respondents, then evaluated and can be de-

scribed with the following hierarchy (in Indone-

sian according to company needs)

Figure 4: Decision Hierarchy

2. Determine priorities (weights) for the criteria

The next step in the AHP process is to determine

the priority (weight) for each standard. This is

done to measure the importance of the require-

ments and then compare with each other. This

process requires a direct assessment of the respon-

dent, namely the distribution of analysts, who are

most knowledgeable about how each of these cri-

teria is applied. As a basis for each weighting, we

use the reference scale set, as the basis for AHP

weighting. And produce it seen in ﬁg. 6.

Figure 5: Pairwise comparison scale (Dweiri et al., 2016)

CONRIST 2019 - International Conferences on Information System and Technology

296

space

Figure 6: Pairwise comparison matrices obtained

The next stage, the normalisation process is de-

rived from the results of the delivery with the

value of SUM. For example the Expeditionary

Fleet on the Expeditionary Fleet 1.00 / 1.65 =

0.60. (See ﬁg.7). After obtaining a complete nor-

malization matrix, the next step is to add priority

vector columns, which are derived from the total

results of each row divided by the number of cri-

teria.

Figure 7: Normalised matrix

So we get the results of the direct assessment of

the respondents in ﬁg. 8 and the effects of priority

vectors.

Figure 8: The results of the original weighting with priority

vector

From the results of the table above it can be seen

that when determining the decision of the destina-

tion warehouse, the support of the expedition ﬂeet

has the highest importance (0.55), followed by the

achievement of distribution (0.20), warehouse re-

quirements (0.15), warehouse capacity (0.07), and

the lowest distance (0.03).

3. Consistency

AHP has set acceptable limits of inconsistency,

namely by calculating the consistency ratio (CR)

which is the result of a comparison of the consis-

tency index (CI) of a matrix with a random matrix

consistency index (RI).

CR = CI/RI (1)

Acceptable inconsistencies are CRs that are less

than or equal to (¡=) 0.10. If the CR is less than

or equal to 0.10 (¡= 0.10), then it can proceed to

the next problem analysis stage, but if more than

0.10, then a revision must be made whether the

assessment is really appropriate. The stage taken

is to multiply the grading matrix with the prior-

ity vector. After the priority vector and weighted

sum are obtained, the next step is to add the result

of division which is the result of the division be-

tween weighted sum and priority vector (see ﬁg.

Figure 9: weighted sum, priority vector, and effect of divi-

sion

λmax = SUM(theresulto f division)

= (5.63 +5.34 + 5.41 +5.03 + 5.1)/5

= 5.302

CI = (λmax–n)/(n–1)

= (5.302 −5)/(5–1)

= 0.075

CR = CI/RI

= 0.075/1.11

= 0.068 (2)

From the above calculation, a consistency ratio

of 0.068 is obtained. This shows that CR is less

than 0.1, so it can be concluded that the assess-

ment/weighting has been done consistently

4. Lower local priorities (preferences) for alterna-

tives

The next step is about how to obtain the rela-

tive priority (preference) of each criterion. This

is done by comparing the criteria against each al-

ternative (pairwise comparison). Here the writer

takes data samples from the distribution analyst

shipping list on April 18, 2019, with ﬂeet police

number B 9499 UEH. This ﬂeet carries only 1

no shipping list, namely 1100/2019/12070, which

contains 51 units of motorcycles. CBS 150 CBX

motorcycles with 8 units of V1J2Q2S2 black

(BK) code, and 43 units of white colour (WH).

To which warehouse will this ﬂeet be allocated?

To determine the solution of the problem sample

above, a sample of questions was conducted by

Decision Support System of Warehouse Allocation using Analytical Hierarchy Process Method

297

the author along with the distribution of analysts

based on predetermined criteria. The results are

shown in ﬁg.10.

Figure 10: Preferences based on the distance of the ware-

house - plant (fast way)

So we get the results of CMG with priority vector

0.29, JTK with priority vector 0.49, and also JTA

with priority vector 0.22. From these ﬁgures, it

can be seen that in terms of plant - warehouse dis-

tance, JTK warehouse is preferred over JTA and

CMG

5. Decrease Overall Priority (Synthesis Model) Af-

ter obtaining local priorities from alternatives that

excel in each criterion, the next step is to calcu-

late the overall preference for each option. From

the global priority, then made the ranking, which

is the result of ordering (order) from the largest to

the smallest.

Figure 11: Ranking results

Because the ranking results have been obtained,

the next stage is not carried out.

4 CONCLUSIONS

Conducted with distribution analysts, it was found

that the proposed system can provide decision support

directly to the company’s PIC Shipping, so that the

allocation process can run in only about 1-2 minutes,

from the previous 9-13 minutes. AHP method used in

the proposed system was able to provide warehouse

allocation decision support with a high degree of ac-

curacy so that the decision automation process could

run well

REFERENCES

Dennis, A., Wixom, B., and Roth, R. (2012). System anal-

ysis & design 5th edition. amerika.

Dweiri, F., Kumar, S., Khan, S. A., and Jain, V. (2016).

Designing an integrated ahp based decision support

system for supplier selection in automotive industry.

Expert Systems with Applications, 62:273–283.

Mu, E. and Pereyra-Rojas, M. (2016). Practical deci-

sion making: an introduction to the Analytic Hierar-

chy Process (AHP) using super decisions, volume 2.

Springer.

Narabin, S. and Boonjing, V. (2016). Selecting students

to a dormitory using ahp. In 2016 13th International

Joint Conference on Computer Science and Software

Engineering (JCSSE), pages 1–5. IEEE.

Pal, K. and Palmer, O. (2000). A decision-support system

for business acquisitions. Decision Support Systems,

27(4):411–429.

Saaty, T. L. (2008). Decision making with the analytic hi-

erarchy process. International journal of services sci-

ences, 1(1):83–98.

Sari, R. E., Meizar, A., Tanjung, D. H., and Nugroho, A. Y.

(2017). Decision making with ahp for selection of

employee. In 2017 5th International Conference on

Cyber and IT Service Management (CITSM), pages 1–

5. IEEE.

Sinesi, S., Altieri, M. G., Marinelli, M., and Dell’Orco, M.

(2017). A multivariate logic decision support system

for optimization of the maritime routes. In 2017 5th

IEEE International Conference on Models and Tech-

nologies for Intelligent Transportation Systems (MT-

ITS), pages 75–79. IEEE.

Susanto, A., Latifah, L., Fitriyani, A., et al. (2017). Deci-

sion support systems design on sharia ﬁnancing using

yager’s fuzzy decision model. In 2017 5th Interna-

tional Conference on Cyber and IT Service Manage-

ment (CITSM), pages 1–4. IEEE.

CONRIST 2019 - International Conferences on Information System and Technology

298