Playing with DISASTER: A Blockchain-Enabled Supply Chain Simulation Platform for Studying Shortages and the Competition for Scarce Resources

Abstract

This chapter explores the potential of distributed ledger technology (DLT) in addressing supply chain shortages and competition for scarce resources. Specifically, we assess the effect of strategic information sharing on supply chain efficiency and the creation of virtual markets to improve supply chain performance. To facilitate this research, we designed a simulation platform called DISASTER (DLT In Sourcing And Strategic Trading Experimental Research), which hosts web-based, dynamic, and customizable supply chain simulations that leverage concepts of blockchain technology, and permit capturing of information regarding players’ ordering strategies and behavioral traits.

In this chapter, we describe the DISASTER platform and discuss two selected DISASTER simulations that probe supply chain retailers’ order behavior: the first investigates the role of information sharing among competing retailers; the second allows for the trading of tokens among competing retailers. In the first simulation, we find that decision makers act more strategically and closer to Nash equilibrium predictions as more information about historical orders of competitors is shared; however, the observed outcome is not invariably an improvement in efficiency as measured by profits across participants. In the second simulation, we observe that initial order quantities remain unchanged as compared to the baseline (non-trading) scenario, despite the possibility to trade on virtual markets; however, over time, more equitable distribution of inventory is achieved, and the supply chain efficiency as measured by profits increases.

Our findings highlight the value of empirical research and management games in shedding light on the role of decision makers’ behavioral characteristics and investigating real-life supply chain challenges and the potential of adopting blockchain-specific capabilities in that space.

Appendix

1.1 A. Comparison of Existing Platforms

Table 2

1.2 Comparison of BWE Simulation Games

Table 3

1.3 C. List of Pre-defined Questions for Eliciting Behavioral Characteristics

Characteristic	Measurement instrument	Reference
Ambiguity aversion	Willingness to pay (WTP) game	Fox and Tversky (1995) and Halevy (2007)
Cognitive abilities	Cognitive Reflection Test (CRT)	Frederick (2005)
Fairness (inequality aversion)	(Hypothetical) Dictator game questionnaire	Forsythe et al. (1994), Fehr and Schmidt (1999) and van Damme et al. (2014)
Loss aversion	Lottery choice task Willingness to accept (WTA) game Willingness to purchase (WTP) game	Kahneman and Tversky (1979), Fehr and Goette (2007) and Gächter et al. (2021)
Positive reciprocity	(Hypothetical) Investment game questionnaire	Berg et al. (1995), Cox (2004), Cooper and Kagel (2016) and Falk et al. (2016)
Negative reciprocity	(Hypothetical) Ultimatum game questionnaire	Forsythe et al. (1994), van Damme et al. (2014) and Falk et al. (2016)
Overconfidence	Questionnaire	Russo and Schoemaker (1992)
Risk preferences	Multiple price list method DOSPERT questionnaire	Holt and Laury (2002) and Blais and Weber (2006)
Trust and trustworthiness	Questionnaire Investment game	Berg et al. (1995), Mayer and Davis (1999) and Falk et al. (2016)

1.4 D. Simulation Instructions for the Information Sharing Among Competing Retailers Game (Scenarios 2–4)

Step 1: Past order observation

You can observe the past order(s) of the other two players against whom you are playing in the current round. Similarly, your past order(s) is (are) visible to the players in your group.

Step 2: Order submission

For your order, you can choose an integer number between 0 and 10,000 units.

Once you have submitted your order, you cannot change it.

Step 3: Supplier stock allocation and cost

The supplier has 120 units available in total, which are divided among all three retailers in your group.

The allocation is determined as follows.

• (a) The system calculates the total order received from the retailers (i.e., the sum of your order and the orders from the other two retailers).

• (b) If the total order is less than or equal to 120, you will receive the number of units you ordered.

• (c) If the total order is greater than 120, you will receive:

  Your allocation = (your order/total order) × 120

Please note that you may receive fractions (decimals) of units.

For example,

• If you order 60 units and the total order from all retailers in your group is 110, then you will receive 60 units. The other two retailers will receive 50 units in total.

• If you order 60 units and the total order from all retailers in your group is 150, then you will receive 48 units [=(60 ÷ 150) × 120]. The other two retailers will receive 72 units in total.

The order cost is $10 per unit. The order cost applies only to units you receive. So, for example, if you receive 60 units then your order cost would be $600 but if you receive just 48 units then your order cost would be $480.

Step 4: Sales and revenues

The number of units you sell is equal to the minimum of (i) the demand from your customers (50 units) and (ii) your supply, or the number of units allocated to you by the supplier (as described in Step 3).

Unsold items are discarded at the end of the round; they are not carried over to the next round. Unsatisfied demand is lost and cannot be backlogged to the next round.

For example,

• if the supplier allocated 40 units to you, then your sales are 40 units (=min(40, 50)) and the unsatisfied demand for 10 units is lost at the end of the round;

• if the supplier allocated 52 units to you, then your sales are 50 units (=min(52, 50)) and the leftover 2 units are lost at the end of the round.

Revenue amounts to $20 per unit. Therefore, if your sales are 40 units then your revenue is $800.

Step 5: Profit

Your profit per round = revenue per round – cost per round.

Your total profit over the entire simulation is the sum of profits per round.

1.5 E. Simulation Instructions for the Trading Tokens Among Competing Retailers Game

Step 1: Observation of your sales price for the current round

Observe the realization of your sales price, which is randomly drawn from the range of $51 to $100 per unit and with an equal likelihood of every integer value (a number without any decimals, such as $51, $63, $95).

The sales price of other retailers is also randomly drawn from the range of $51 to $100 per unit.

Sales prices are independent between rounds and across retailers.

Step 2: Submit your order to the supplier

For your order, choose an integer number between 0 and 400 units.

You can use the simulation tool to support your decision on how many tokens to order from the supplier. This tool will provide you with the expected pre-trade profit after you enter your estimated demand and specified order quantity.

Step 3: Demand realization

Your customer demand is randomly drawn from the range of 0 to 200 units, with an equal likelihood of every integer value (a number without any decimals, e.g., 13, 105, 186).

Other players face their own levels of customer demand (i.e., you are not competing for the same customers), which is also randomly drawn from the range of 0 to 200 units.

Customer demands are independent between rounds and across retailers.

Step 4: Cost, sales, and revenue

You can observe your pre-trade profit projection. If you do not submit any trading orders, then that projection would be your final profit in this round. The following text describes how the projected pre-trade profit is calculated.

The order cost is $10 per token. For instance, if you order 130 tokens then your cost is $1300.

The number of units you can sell to your customers is equal to the minimum of (i) the demand from your customers (see Step 3) and (ii) how many tokens of the supplier’s capacity that you hold (see Step 2). Suppose, for example, that you hold 130 tokens and that your customer demand is 90 units; in that case, your potential sales quantity is 90 units (=min(130, 90)).

Your projected pre-trade revenue is equal to the sales price multiplied by the sales quantity. So if your sales price is $60 and you sell 90 units, then your projected pre-trade revenue is $5400.

Finally: Projected pre-trade profit = projected pre-trade revenue – order cost.

Step 5: Perform trades with other retailers in your group

You can change how many tokens you hold by trading with other retailers.

To trade, you specify whether you want to buy or sell tokens, the quantity, and the price. You can sell all tokens that you hold (even if you can then not fulfill your customer demand as the result). You can submit multiple trade orders per round, up to a maximum of five.

At the end of the trading period, the market clears given all orders that the retailers in your group have submitted.

Trading and market-clearing processes (see table below)

• A. Sell orders are ranked in price from lowest to highest and buy orders in price from highest to lowest.

• B. Units are matched in buy and sell orders whenever the buy price is greater than the sell price of the matched units.

• C. The average of the lowest buy price and the highest sell price at which the last match happens is the market-clearing price: all buy orders pay this price and all sell orders receive this price.

Consider the following example. The steps just described are marked by A, B, and C in the table.

After the trading phase is completed, you sell to your customers using the new number of tokens you have.

This marks the end of the round, and the final profits for this round are then calculated. Your total profit over the entire simulation is the sum of profits per round.

Any unsold tokens are voided at the end of the round; they are not carried over to the next round.

Unsatisfied customer demand is lost, and it cannot be backlogged to the next round.